Commit
•
f17a09c
1
Parent(s):
98878b5
Model save
Browse files- README.md +68 -0
- all_results.json +9 -0
- generation_config.json +8 -0
- model-00001-of-00002.safetensors +1 -1
- model-00002-of-00002.safetensors +1 -1
- train_results.json +9 -0
- trainer_state.json +2533 -0
README.md
ADDED
|
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: gemma
|
| 3 |
+
base_model: google/gemma-2-2b
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: easylm-helpsteer-sft-gemma-2-2b
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# easylm-helpsteer-sft-gemma-2-2b
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [google/gemma-2-2b](https://huggingface.co/google/gemma-2-2b) on an unknown dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 1.1501
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 3e-06
|
| 40 |
+
- train_batch_size: 1
|
| 41 |
+
- eval_batch_size: 1
|
| 42 |
+
- seed: 42
|
| 43 |
+
- distributed_type: multi-GPU
|
| 44 |
+
- num_devices: 4
|
| 45 |
+
- total_train_batch_size: 4
|
| 46 |
+
- total_eval_batch_size: 4
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: cosine
|
| 49 |
+
- num_epochs: 2
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 54 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 55 |
+
| 1.1281 | 0.2864 | 500 | 1.1230 |
|
| 56 |
+
| 1.183 | 0.5727 | 1000 | 1.1017 |
|
| 57 |
+
| 1.1009 | 0.8591 | 1500 | 1.0869 |
|
| 58 |
+
| 0.6901 | 1.1455 | 2000 | 1.1486 |
|
| 59 |
+
| 0.7377 | 1.4318 | 2500 | 1.1488 |
|
| 60 |
+
| 0.6947 | 1.7182 | 3000 | 1.1501 |
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
### Framework versions
|
| 64 |
+
|
| 65 |
+
- Transformers 4.43.3
|
| 66 |
+
- Pytorch 2.3.0+cu121
|
| 67 |
+
- Datasets 2.20.0
|
| 68 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"total_flos": 8.351047625159475e+16,
|
| 4 |
+
"train_loss": 0.9289274918937465,
|
| 5 |
+
"train_runtime": 3314.9598,
|
| 6 |
+
"train_samples": 6981,
|
| 7 |
+
"train_samples_per_second": 4.212,
|
| 8 |
+
"train_steps_per_second": 1.053
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"cache_implementation": "hybrid",
|
| 5 |
+
"eos_token_id": 1,
|
| 6 |
+
"pad_token_id": 0,
|
| 7 |
+
"transformers_version": "4.43.3"
|
| 8 |
+
}
|
model-00001-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4988025760
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a7d265a5e05ac710dcf02ecd058e8efdcb6c973ac5c947d4d8349a3cf39b702
|
| 3 |
size 4988025760
|
model-00002-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1420339880
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7f20aad536787818e15aa114ffc2bd29391c95d69086797b833de0c57b4d4e6f
|
| 3 |
size 1420339880
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"total_flos": 8.351047625159475e+16,
|
| 4 |
+
"train_loss": 0.9289274918937465,
|
| 5 |
+
"train_runtime": 3314.9598,
|
| 6 |
+
"train_samples": 6981,
|
| 7 |
+
"train_samples_per_second": 4.212,
|
| 8 |
+
"train_steps_per_second": 1.053
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2533 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 3492,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0057273768613974796,
|
| 13 |
+
"grad_norm": 16.379655838012695,
|
| 14 |
+
"learning_rate": 2.9999392970762296e-06,
|
| 15 |
+
"loss": 1.5958,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.011454753722794959,
|
| 20 |
+
"grad_norm": 13.56667423248291,
|
| 21 |
+
"learning_rate": 2.999757193218044e-06,
|
| 22 |
+
"loss": 1.3928,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.01718213058419244,
|
| 27 |
+
"grad_norm": 10.130936622619629,
|
| 28 |
+
"learning_rate": 2.999453703164426e-06,
|
| 29 |
+
"loss": 1.2605,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.022909507445589918,
|
| 34 |
+
"grad_norm": 10.301068305969238,
|
| 35 |
+
"learning_rate": 2.9990288514790206e-06,
|
| 36 |
+
"loss": 1.3193,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0286368843069874,
|
| 41 |
+
"grad_norm": 8.71345043182373,
|
| 42 |
+
"learning_rate": 2.9984826725481468e-06,
|
| 43 |
+
"loss": 1.1754,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.03436426116838488,
|
| 48 |
+
"grad_norm": 12.313104629516602,
|
| 49 |
+
"learning_rate": 2.9978152105780155e-06,
|
| 50 |
+
"loss": 1.3135,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.04009163802978236,
|
| 55 |
+
"grad_norm": 12.800464630126953,
|
| 56 |
+
"learning_rate": 2.9970265195911502e-06,
|
| 57 |
+
"loss": 1.2516,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.045819014891179836,
|
| 62 |
+
"grad_norm": 10.99730396270752,
|
| 63 |
+
"learning_rate": 2.9961166634220168e-06,
|
| 64 |
+
"loss": 1.203,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.05154639175257732,
|
| 69 |
+
"grad_norm": 13.54122257232666,
|
| 70 |
+
"learning_rate": 2.9950857157118544e-06,
|
| 71 |
+
"loss": 1.2237,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0572737686139748,
|
| 76 |
+
"grad_norm": 12.151561737060547,
|
| 77 |
+
"learning_rate": 2.9939337599027165e-06,
|
| 78 |
+
"loss": 1.2715,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.06300114547537228,
|
| 83 |
+
"grad_norm": 19.66217803955078,
|
| 84 |
+
"learning_rate": 2.992660889230718e-06,
|
| 85 |
+
"loss": 1.2601,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.06872852233676977,
|
| 90 |
+
"grad_norm": 9.985322952270508,
|
| 91 |
+
"learning_rate": 2.9912672067184863e-06,
|
| 92 |
+
"loss": 1.1591,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.07445589919816724,
|
| 97 |
+
"grad_norm": 10.843229293823242,
|
| 98 |
+
"learning_rate": 2.9897528251668274e-06,
|
| 99 |
+
"loss": 1.1064,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.08018327605956473,
|
| 104 |
+
"grad_norm": 13.866814613342285,
|
| 105 |
+
"learning_rate": 2.9881178671455905e-06,
|
| 106 |
+
"loss": 1.1918,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.0859106529209622,
|
| 111 |
+
"grad_norm": 13.748754501342773,
|
| 112 |
+
"learning_rate": 2.9863624649837522e-06,
|
| 113 |
+
"loss": 1.2726,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.09163802978235967,
|
| 118 |
+
"grad_norm": 9.494117736816406,
|
| 119 |
+
"learning_rate": 2.9844867607587037e-06,
|
| 120 |
+
"loss": 1.1832,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.09736540664375716,
|
| 125 |
+
"grad_norm": 9.150392532348633,
|
| 126 |
+
"learning_rate": 2.982490906284753e-06,
|
| 127 |
+
"loss": 1.1274,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.10309278350515463,
|
| 132 |
+
"grad_norm": 11.971685409545898,
|
| 133 |
+
"learning_rate": 2.980375063100836e-06,
|
| 134 |
+
"loss": 1.1399,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.10882016036655212,
|
| 139 |
+
"grad_norm": 8.963068962097168,
|
| 140 |
+
"learning_rate": 2.9781394024574426e-06,
|
| 141 |
+
"loss": 1.1726,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.1145475372279496,
|
| 146 |
+
"grad_norm": 9.128679275512695,
|
| 147 |
+
"learning_rate": 2.975784105302756e-06,
|
| 148 |
+
"loss": 1.1972,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.12027491408934708,
|
| 153 |
+
"grad_norm": 12.733792304992676,
|
| 154 |
+
"learning_rate": 2.973309362268008e-06,
|
| 155 |
+
"loss": 1.1732,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.12600229095074456,
|
| 160 |
+
"grad_norm": 10.524323463439941,
|
| 161 |
+
"learning_rate": 2.970715373652049e-06,
|
| 162 |
+
"loss": 1.2613,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.13172966781214204,
|
| 167 |
+
"grad_norm": 7.896759510040283,
|
| 168 |
+
"learning_rate": 2.9680023494051364e-06,
|
| 169 |
+
"loss": 1.1437,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.13745704467353953,
|
| 174 |
+
"grad_norm": 7.099399089813232,
|
| 175 |
+
"learning_rate": 2.9651705091119422e-06,
|
| 176 |
+
"loss": 1.0787,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.143184421534937,
|
| 181 |
+
"grad_norm": 9.507390975952148,
|
| 182 |
+
"learning_rate": 2.962220081973781e-06,
|
| 183 |
+
"loss": 1.0828,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.14891179839633448,
|
| 188 |
+
"grad_norm": 10.189469337463379,
|
| 189 |
+
"learning_rate": 2.959151306790057e-06,
|
| 190 |
+
"loss": 1.1324,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.15463917525773196,
|
| 195 |
+
"grad_norm": 10.902261734008789,
|
| 196 |
+
"learning_rate": 2.955964431938939e-06,
|
| 197 |
+
"loss": 1.2347,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.16036655211912945,
|
| 202 |
+
"grad_norm": 10.873947143554688,
|
| 203 |
+
"learning_rate": 2.952659715357254e-06,
|
| 204 |
+
"loss": 1.1749,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.1660939289805269,
|
| 209 |
+
"grad_norm": 8.8741455078125,
|
| 210 |
+
"learning_rate": 2.9492374245196148e-06,
|
| 211 |
+
"loss": 1.2453,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.1718213058419244,
|
| 216 |
+
"grad_norm": 9.256343841552734,
|
| 217 |
+
"learning_rate": 2.9456978364167667e-06,
|
| 218 |
+
"loss": 1.2202,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.1775486827033219,
|
| 223 |
+
"grad_norm": 13.664665222167969,
|
| 224 |
+
"learning_rate": 2.9420412375331737e-06,
|
| 225 |
+
"loss": 1.1165,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.18327605956471935,
|
| 230 |
+
"grad_norm": 12.664402961730957,
|
| 231 |
+
"learning_rate": 2.938267923823825e-06,
|
| 232 |
+
"loss": 1.1933,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.18900343642611683,
|
| 237 |
+
"grad_norm": 10.518999099731445,
|
| 238 |
+
"learning_rate": 2.934378200690288e-06,
|
| 239 |
+
"loss": 1.1306,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.19473081328751432,
|
| 244 |
+
"grad_norm": 11.800782203674316,
|
| 245 |
+
"learning_rate": 2.9303723829559846e-06,
|
| 246 |
+
"loss": 1.1013,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.2004581901489118,
|
| 251 |
+
"grad_norm": 13.071691513061523,
|
| 252 |
+
"learning_rate": 2.9262507948407135e-06,
|
| 253 |
+
"loss": 1.2304,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.20618556701030927,
|
| 258 |
+
"grad_norm": 9.5667142868042,
|
| 259 |
+
"learning_rate": 2.9220137699344057e-06,
|
| 260 |
+
"loss": 1.1543,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.21191294387170675,
|
| 265 |
+
"grad_norm": 11.27755069732666,
|
| 266 |
+
"learning_rate": 2.917661651170129e-06,
|
| 267 |
+
"loss": 1.1647,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.21764032073310424,
|
| 272 |
+
"grad_norm": 12.141744613647461,
|
| 273 |
+
"learning_rate": 2.913194790796327e-06,
|
| 274 |
+
"loss": 1.1378,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.22336769759450173,
|
| 279 |
+
"grad_norm": 14.818400382995605,
|
| 280 |
+
"learning_rate": 2.9086135503483137e-06,
|
| 281 |
+
"loss": 1.1353,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.2290950744558992,
|
| 286 |
+
"grad_norm": 10.42490005493164,
|
| 287 |
+
"learning_rate": 2.903918300619008e-06,
|
| 288 |
+
"loss": 1.2457,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.23482245131729668,
|
| 293 |
+
"grad_norm": 11.577737808227539,
|
| 294 |
+
"learning_rate": 2.8991094216289254e-06,
|
| 295 |
+
"loss": 1.1824,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.24054982817869416,
|
| 300 |
+
"grad_norm": 9.93144416809082,
|
| 301 |
+
"learning_rate": 2.8941873025954193e-06,
|
| 302 |
+
"loss": 1.2134,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.24627720504009165,
|
| 307 |
+
"grad_norm": 8.588950157165527,
|
| 308 |
+
"learning_rate": 2.8891523419011774e-06,
|
| 309 |
+
"loss": 1.2043,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.2520045819014891,
|
| 314 |
+
"grad_norm": 9.073149681091309,
|
| 315 |
+
"learning_rate": 2.884004947061981e-06,
|
| 316 |
+
"loss": 1.2291,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.25773195876288657,
|
| 321 |
+
"grad_norm": 7.688851833343506,
|
| 322 |
+
"learning_rate": 2.8787455346937186e-06,
|
| 323 |
+
"loss": 1.2401,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.2634593356242841,
|
| 328 |
+
"grad_norm": 10.9544677734375,
|
| 329 |
+
"learning_rate": 2.873374530478667e-06,
|
| 330 |
+
"loss": 1.2467,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.26918671248568155,
|
| 335 |
+
"grad_norm": 9.605713844299316,
|
| 336 |
+
"learning_rate": 2.8678923691310392e-06,
|
| 337 |
+
"loss": 1.1678,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.27491408934707906,
|
| 342 |
+
"grad_norm": 11.065582275390625,
|
| 343 |
+
"learning_rate": 2.8622994943617984e-06,
|
| 344 |
+
"loss": 1.1364,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.2806414662084765,
|
| 349 |
+
"grad_norm": 9.69209098815918,
|
| 350 |
+
"learning_rate": 2.8565963588427457e-06,
|
| 351 |
+
"loss": 1.0762,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.286368843069874,
|
| 356 |
+
"grad_norm": 10.874582290649414,
|
| 357 |
+
"learning_rate": 2.8507834241698817e-06,
|
| 358 |
+
"loss": 1.1281,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.286368843069874,
|
| 363 |
+
"eval_loss": 1.1230082511901855,
|
| 364 |
+
"eval_runtime": 5.3115,
|
| 365 |
+
"eval_samples_per_second": 67.778,
|
| 366 |
+
"eval_steps_per_second": 16.944,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.2920962199312715,
|
| 371 |
+
"grad_norm": 16.209104537963867,
|
| 372 |
+
"learning_rate": 2.844861160826047e-06,
|
| 373 |
+
"loss": 1.158,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.29782359679266895,
|
| 378 |
+
"grad_norm": 11.166873931884766,
|
| 379 |
+
"learning_rate": 2.8388300481428423e-06,
|
| 380 |
+
"loss": 1.0916,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.3035509736540664,
|
| 385 |
+
"grad_norm": 11.725543022155762,
|
| 386 |
+
"learning_rate": 2.8326905742618318e-06,
|
| 387 |
+
"loss": 1.2,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.30927835051546393,
|
| 392 |
+
"grad_norm": 10.414682388305664,
|
| 393 |
+
"learning_rate": 2.8264432360950353e-06,
|
| 394 |
+
"loss": 1.1615,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.3150057273768614,
|
| 399 |
+
"grad_norm": 9.764941215515137,
|
| 400 |
+
"learning_rate": 2.8200885392847106e-06,
|
| 401 |
+
"loss": 1.175,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.3207331042382589,
|
| 406 |
+
"grad_norm": 16.335607528686523,
|
| 407 |
+
"learning_rate": 2.8136269981624244e-06,
|
| 408 |
+
"loss": 1.0646,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.32646048109965636,
|
| 413 |
+
"grad_norm": 8.465163230895996,
|
| 414 |
+
"learning_rate": 2.8070591357074282e-06,
|
| 415 |
+
"loss": 1.2205,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.3321878579610538,
|
| 420 |
+
"grad_norm": 11.596179008483887,
|
| 421 |
+
"learning_rate": 2.800385483504327e-06,
|
| 422 |
+
"loss": 1.2421,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.33791523482245134,
|
| 427 |
+
"grad_norm": 11.226552963256836,
|
| 428 |
+
"learning_rate": 2.7936065817000565e-06,
|
| 429 |
+
"loss": 1.1326,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.3436426116838488,
|
| 434 |
+
"grad_norm": 11.108230590820312,
|
| 435 |
+
"learning_rate": 2.7867229789601615e-06,
|
| 436 |
+
"loss": 1.2161,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.34936998854524626,
|
| 441 |
+
"grad_norm": 12.34344482421875,
|
| 442 |
+
"learning_rate": 2.7797352324243925e-06,
|
| 443 |
+
"loss": 1.149,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.3550973654066438,
|
| 448 |
+
"grad_norm": 7.934173583984375,
|
| 449 |
+
"learning_rate": 2.772643907661609e-06,
|
| 450 |
+
"loss": 1.1019,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.36082474226804123,
|
| 455 |
+
"grad_norm": 7.608425140380859,
|
| 456 |
+
"learning_rate": 2.765449578624007e-06,
|
| 457 |
+
"loss": 1.0918,
|
| 458 |
+
"step": 630
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.3665521191294387,
|
| 462 |
+
"grad_norm": 11.046185493469238,
|
| 463 |
+
"learning_rate": 2.7581528276006626e-06,
|
| 464 |
+
"loss": 1.0596,
|
| 465 |
+
"step": 640
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.3722794959908362,
|
| 469 |
+
"grad_norm": 12.330936431884766,
|
| 470 |
+
"learning_rate": 2.7507542451704043e-06,
|
| 471 |
+
"loss": 1.282,
|
| 472 |
+
"step": 650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.37800687285223367,
|
| 476 |
+
"grad_norm": 14.420117378234863,
|
| 477 |
+
"learning_rate": 2.743254430154012e-06,
|
| 478 |
+
"loss": 1.215,
|
| 479 |
+
"step": 660
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.3837342497136312,
|
| 483 |
+
"grad_norm": 10.501232147216797,
|
| 484 |
+
"learning_rate": 2.7356539895657513e-06,
|
| 485 |
+
"loss": 1.1786,
|
| 486 |
+
"step": 670
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.38946162657502864,
|
| 490 |
+
"grad_norm": 10.232950210571289,
|
| 491 |
+
"learning_rate": 2.727953538564244e-06,
|
| 492 |
+
"loss": 1.1201,
|
| 493 |
+
"step": 680
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.3951890034364261,
|
| 497 |
+
"grad_norm": 13.221607208251953,
|
| 498 |
+
"learning_rate": 2.7201537004026756e-06,
|
| 499 |
+
"loss": 1.1669,
|
| 500 |
+
"step": 690
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.4009163802978236,
|
| 504 |
+
"grad_norm": 9.83636474609375,
|
| 505 |
+
"learning_rate": 2.7122551063783555e-06,
|
| 506 |
+
"loss": 1.2779,
|
| 507 |
+
"step": 700
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.4066437571592211,
|
| 511 |
+
"grad_norm": 10.758344650268555,
|
| 512 |
+
"learning_rate": 2.704258395781618e-06,
|
| 513 |
+
"loss": 1.1545,
|
| 514 |
+
"step": 710
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.41237113402061853,
|
| 518 |
+
"grad_norm": 15.233297348022461,
|
| 519 |
+
"learning_rate": 2.696164215844081e-06,
|
| 520 |
+
"loss": 1.0843,
|
| 521 |
+
"step": 720
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.41809851088201605,
|
| 525 |
+
"grad_norm": 9.102065086364746,
|
| 526 |
+
"learning_rate": 2.687973221686263e-06,
|
| 527 |
+
"loss": 1.0776,
|
| 528 |
+
"step": 730
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.4238258877434135,
|
| 532 |
+
"grad_norm": 9.260092735290527,
|
| 533 |
+
"learning_rate": 2.679686076264554e-06,
|
| 534 |
+
"loss": 1.1087,
|
| 535 |
+
"step": 740
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.42955326460481097,
|
| 539 |
+
"grad_norm": 19.753076553344727,
|
| 540 |
+
"learning_rate": 2.6713034503175643e-06,
|
| 541 |
+
"loss": 1.1055,
|
| 542 |
+
"step": 750
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.4352806414662085,
|
| 546 |
+
"grad_norm": 7.540209770202637,
|
| 547 |
+
"learning_rate": 2.662826022311832e-06,
|
| 548 |
+
"loss": 1.1342,
|
| 549 |
+
"step": 760
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.44100801832760594,
|
| 553 |
+
"grad_norm": 12.86109733581543,
|
| 554 |
+
"learning_rate": 2.654254478386912e-06,
|
| 555 |
+
"loss": 1.186,
|
| 556 |
+
"step": 770
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.44673539518900346,
|
| 560 |
+
"grad_norm": 9.657477378845215,
|
| 561 |
+
"learning_rate": 2.6455895122998405e-06,
|
| 562 |
+
"loss": 1.1337,
|
| 563 |
+
"step": 780
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.4524627720504009,
|
| 567 |
+
"grad_norm": 11.697619438171387,
|
| 568 |
+
"learning_rate": 2.636831825368986e-06,
|
| 569 |
+
"loss": 1.1123,
|
| 570 |
+
"step": 790
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.4581901489117984,
|
| 574 |
+
"grad_norm": 9.250412940979004,
|
| 575 |
+
"learning_rate": 2.627982126417284e-06,
|
| 576 |
+
"loss": 1.0457,
|
| 577 |
+
"step": 800
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.4639175257731959,
|
| 581 |
+
"grad_norm": 12.023123741149902,
|
| 582 |
+
"learning_rate": 2.619041131714869e-06,
|
| 583 |
+
"loss": 1.0837,
|
| 584 |
+
"step": 810
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.46964490263459335,
|
| 588 |
+
"grad_norm": 8.030861854553223,
|
| 589 |
+
"learning_rate": 2.6100095649211017e-06,
|
| 590 |
+
"loss": 1.1827,
|
| 591 |
+
"step": 820
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.4753722794959908,
|
| 595 |
+
"grad_norm": 10.623430252075195,
|
| 596 |
+
"learning_rate": 2.600888157025995e-06,
|
| 597 |
+
"loss": 1.0982,
|
| 598 |
+
"step": 830
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.48109965635738833,
|
| 602 |
+
"grad_norm": 8.662152290344238,
|
| 603 |
+
"learning_rate": 2.591677646291054e-06,
|
| 604 |
+
"loss": 1.2227,
|
| 605 |
+
"step": 840
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.4868270332187858,
|
| 609 |
+
"grad_norm": 11.325638771057129,
|
| 610 |
+
"learning_rate": 2.5823787781895197e-06,
|
| 611 |
+
"loss": 0.9943,
|
| 612 |
+
"step": 850
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.4925544100801833,
|
| 616 |
+
"grad_norm": 9.88836669921875,
|
| 617 |
+
"learning_rate": 2.5729923053460347e-06,
|
| 618 |
+
"loss": 1.2253,
|
| 619 |
+
"step": 860
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.49828178694158076,
|
| 623 |
+
"grad_norm": 11.135693550109863,
|
| 624 |
+
"learning_rate": 2.5635189874757255e-06,
|
| 625 |
+
"loss": 1.2906,
|
| 626 |
+
"step": 870
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.5040091638029782,
|
| 630 |
+
"grad_norm": 8.991135597229004,
|
| 631 |
+
"learning_rate": 2.553959591322716e-06,
|
| 632 |
+
"loss": 1.0484,
|
| 633 |
+
"step": 880
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.5097365406643757,
|
| 637 |
+
"grad_norm": 8.861202239990234,
|
| 638 |
+
"learning_rate": 2.544314890598067e-06,
|
| 639 |
+
"loss": 1.0301,
|
| 640 |
+
"step": 890
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.5154639175257731,
|
| 644 |
+
"grad_norm": 13.207406997680664,
|
| 645 |
+
"learning_rate": 2.5345856659171565e-06,
|
| 646 |
+
"loss": 1.2542,
|
| 647 |
+
"step": 900
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.5211912943871707,
|
| 651 |
+
"grad_norm": 9.178120613098145,
|
| 652 |
+
"learning_rate": 2.5247727047364956e-06,
|
| 653 |
+
"loss": 1.085,
|
| 654 |
+
"step": 910
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.5269186712485682,
|
| 658 |
+
"grad_norm": 10.854491233825684,
|
| 659 |
+
"learning_rate": 2.5148768012899977e-06,
|
| 660 |
+
"loss": 1.2201,
|
| 661 |
+
"step": 920
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.5326460481099656,
|
| 665 |
+
"grad_norm": 10.438577651977539,
|
| 666 |
+
"learning_rate": 2.5048987565246934e-06,
|
| 667 |
+
"loss": 1.1778,
|
| 668 |
+
"step": 930
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.5383734249713631,
|
| 672 |
+
"grad_norm": 10.084646224975586,
|
| 673 |
+
"learning_rate": 2.4948393780359025e-06,
|
| 674 |
+
"loss": 1.1588,
|
| 675 |
+
"step": 940
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.5441008018327605,
|
| 679 |
+
"grad_norm": 10.475250244140625,
|
| 680 |
+
"learning_rate": 2.484699480001873e-06,
|
| 681 |
+
"loss": 1.138,
|
| 682 |
+
"step": 950
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.5498281786941581,
|
| 686 |
+
"grad_norm": 11.831788063049316,
|
| 687 |
+
"learning_rate": 2.4744798831178817e-06,
|
| 688 |
+
"loss": 1.1661,
|
| 689 |
+
"step": 960
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.5555555555555556,
|
| 693 |
+
"grad_norm": 11.067156791687012,
|
| 694 |
+
"learning_rate": 2.464181414529809e-06,
|
| 695 |
+
"loss": 1.047,
|
| 696 |
+
"step": 970
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.561282932416953,
|
| 700 |
+
"grad_norm": 12.043890953063965,
|
| 701 |
+
"learning_rate": 2.4538049077671936e-06,
|
| 702 |
+
"loss": 1.2494,
|
| 703 |
+
"step": 980
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.5670103092783505,
|
| 707 |
+
"grad_norm": 12.70505142211914,
|
| 708 |
+
"learning_rate": 2.4433512026757667e-06,
|
| 709 |
+
"loss": 1.1073,
|
| 710 |
+
"step": 990
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.572737686139748,
|
| 714 |
+
"grad_norm": 12.398516654968262,
|
| 715 |
+
"learning_rate": 2.4328211453494805e-06,
|
| 716 |
+
"loss": 1.183,
|
| 717 |
+
"step": 1000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.572737686139748,
|
| 721 |
+
"eval_loss": 1.1016738414764404,
|
| 722 |
+
"eval_runtime": 5.3796,
|
| 723 |
+
"eval_samples_per_second": 66.92,
|
| 724 |
+
"eval_steps_per_second": 16.73,
|
| 725 |
+
"step": 1000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.5784650630011455,
|
| 729 |
+
"grad_norm": 12.058919906616211,
|
| 730 |
+
"learning_rate": 2.4222155880620232e-06,
|
| 731 |
+
"loss": 1.1006,
|
| 732 |
+
"step": 1010
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.584192439862543,
|
| 736 |
+
"grad_norm": 9.806292533874512,
|
| 737 |
+
"learning_rate": 2.4115353891978432e-06,
|
| 738 |
+
"loss": 1.1667,
|
| 739 |
+
"step": 1020
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 0.5899198167239404,
|
| 743 |
+
"grad_norm": 8.650660514831543,
|
| 744 |
+
"learning_rate": 2.4007814131826705e-06,
|
| 745 |
+
"loss": 1.1308,
|
| 746 |
+
"step": 1030
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.5956471935853379,
|
| 750 |
+
"grad_norm": 11.571444511413574,
|
| 751 |
+
"learning_rate": 2.3899545304135533e-06,
|
| 752 |
+
"loss": 1.2669,
|
| 753 |
+
"step": 1040
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.6013745704467354,
|
| 757 |
+
"grad_norm": 10.114933013916016,
|
| 758 |
+
"learning_rate": 2.3790556171884107e-06,
|
| 759 |
+
"loss": 1.1049,
|
| 760 |
+
"step": 1050
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.6071019473081328,
|
| 764 |
+
"grad_norm": 11.771845817565918,
|
| 765 |
+
"learning_rate": 2.3680855556351075e-06,
|
| 766 |
+
"loss": 1.1428,
|
| 767 |
+
"step": 1060
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.6128293241695304,
|
| 771 |
+
"grad_norm": 10.653964042663574,
|
| 772 |
+
"learning_rate": 2.357045233640057e-06,
|
| 773 |
+
"loss": 1.1645,
|
| 774 |
+
"step": 1070
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.6185567010309279,
|
| 778 |
+
"grad_norm": 7.707538604736328,
|
| 779 |
+
"learning_rate": 2.34593554477636e-06,
|
| 780 |
+
"loss": 1.0534,
|
| 781 |
+
"step": 1080
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.6242840778923253,
|
| 785 |
+
"grad_norm": 13.67249870300293,
|
| 786 |
+
"learning_rate": 2.3347573882314764e-06,
|
| 787 |
+
"loss": 1.1389,
|
| 788 |
+
"step": 1090
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.6300114547537228,
|
| 792 |
+
"grad_norm": 11.033611297607422,
|
| 793 |
+
"learning_rate": 2.323511668734453e-06,
|
| 794 |
+
"loss": 1.2522,
|
| 795 |
+
"step": 1100
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 0.6357388316151202,
|
| 799 |
+
"grad_norm": 10.89904499053955,
|
| 800 |
+
"learning_rate": 2.312199296482695e-06,
|
| 801 |
+
"loss": 1.2145,
|
| 802 |
+
"step": 1110
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 0.6414662084765178,
|
| 806 |
+
"grad_norm": 15.723979949951172,
|
| 807 |
+
"learning_rate": 2.3008211870682958e-06,
|
| 808 |
+
"loss": 1.1635,
|
| 809 |
+
"step": 1120
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.6471935853379153,
|
| 813 |
+
"grad_norm": 9.854010581970215,
|
| 814 |
+
"learning_rate": 2.2893782614039325e-06,
|
| 815 |
+
"loss": 1.1548,
|
| 816 |
+
"step": 1130
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.6529209621993127,
|
| 820 |
+
"grad_norm": 7.649772644042969,
|
| 821 |
+
"learning_rate": 2.277871445648332e-06,
|
| 822 |
+
"loss": 1.1,
|
| 823 |
+
"step": 1140
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 0.6586483390607102,
|
| 827 |
+
"grad_norm": 10.99400806427002,
|
| 828 |
+
"learning_rate": 2.266301671131307e-06,
|
| 829 |
+
"loss": 1.0816,
|
| 830 |
+
"step": 1150
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.6643757159221076,
|
| 834 |
+
"grad_norm": 10.330283164978027,
|
| 835 |
+
"learning_rate": 2.2546698742783778e-06,
|
| 836 |
+
"loss": 1.0854,
|
| 837 |
+
"step": 1160
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 0.6701030927835051,
|
| 841 |
+
"grad_norm": 10.694668769836426,
|
| 842 |
+
"learning_rate": 2.2429769965349817e-06,
|
| 843 |
+
"loss": 1.1252,
|
| 844 |
+
"step": 1170
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 0.6758304696449027,
|
| 848 |
+
"grad_norm": 8.67434310913086,
|
| 849 |
+
"learning_rate": 2.231223984290273e-06,
|
| 850 |
+
"loss": 1.2776,
|
| 851 |
+
"step": 1180
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.6815578465063001,
|
| 855 |
+
"grad_norm": 16.908462524414062,
|
| 856 |
+
"learning_rate": 2.219411788800528e-06,
|
| 857 |
+
"loss": 1.1514,
|
| 858 |
+
"step": 1190
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.6872852233676976,
|
| 862 |
+
"grad_norm": 9.732044219970703,
|
| 863 |
+
"learning_rate": 2.2075413661121492e-06,
|
| 864 |
+
"loss": 1.1616,
|
| 865 |
+
"step": 1200
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.693012600229095,
|
| 869 |
+
"grad_norm": 12.07868766784668,
|
| 870 |
+
"learning_rate": 2.1956136769842885e-06,
|
| 871 |
+
"loss": 1.0862,
|
| 872 |
+
"step": 1210
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.6987399770904925,
|
| 876 |
+
"grad_norm": 15.745189666748047,
|
| 877 |
+
"learning_rate": 2.1836296868110836e-06,
|
| 878 |
+
"loss": 1.172,
|
| 879 |
+
"step": 1220
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 0.7044673539518901,
|
| 883 |
+
"grad_norm": 7.780303478240967,
|
| 884 |
+
"learning_rate": 2.1715903655435243e-06,
|
| 885 |
+
"loss": 1.1839,
|
| 886 |
+
"step": 1230
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 0.7101947308132875,
|
| 890 |
+
"grad_norm": 10.862656593322754,
|
| 891 |
+
"learning_rate": 2.1594966876109456e-06,
|
| 892 |
+
"loss": 1.2164,
|
| 893 |
+
"step": 1240
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 0.715922107674685,
|
| 897 |
+
"grad_norm": 11.16341495513916,
|
| 898 |
+
"learning_rate": 2.1473496318421603e-06,
|
| 899 |
+
"loss": 1.1294,
|
| 900 |
+
"step": 1250
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 0.7216494845360825,
|
| 904 |
+
"grad_norm": 6.848244667053223,
|
| 905 |
+
"learning_rate": 2.135150181386236e-06,
|
| 906 |
+
"loss": 1.1519,
|
| 907 |
+
"step": 1260
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 0.7273768613974799,
|
| 911 |
+
"grad_norm": 11.13603401184082,
|
| 912 |
+
"learning_rate": 2.122899323632919e-06,
|
| 913 |
+
"loss": 1.1782,
|
| 914 |
+
"step": 1270
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 0.7331042382588774,
|
| 918 |
+
"grad_norm": 7.1757402420043945,
|
| 919 |
+
"learning_rate": 2.1105980501327234e-06,
|
| 920 |
+
"loss": 1.0794,
|
| 921 |
+
"step": 1280
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 0.738831615120275,
|
| 925 |
+
"grad_norm": 8.15458869934082,
|
| 926 |
+
"learning_rate": 2.0982473565166727e-06,
|
| 927 |
+
"loss": 1.0031,
|
| 928 |
+
"step": 1290
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 0.7445589919816724,
|
| 932 |
+
"grad_norm": 11.527698516845703,
|
| 933 |
+
"learning_rate": 2.0858482424157164e-06,
|
| 934 |
+
"loss": 1.1078,
|
| 935 |
+
"step": 1300
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 0.7502863688430699,
|
| 939 |
+
"grad_norm": 7.272953510284424,
|
| 940 |
+
"learning_rate": 2.073401711379827e-06,
|
| 941 |
+
"loss": 1.2212,
|
| 942 |
+
"step": 1310
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 0.7560137457044673,
|
| 946 |
+
"grad_norm": 8.701218605041504,
|
| 947 |
+
"learning_rate": 2.060908770796769e-06,
|
| 948 |
+
"loss": 1.1787,
|
| 949 |
+
"step": 1320
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 0.7617411225658648,
|
| 953 |
+
"grad_norm": 8.603687286376953,
|
| 954 |
+
"learning_rate": 2.0483704318105704e-06,
|
| 955 |
+
"loss": 1.0969,
|
| 956 |
+
"step": 1330
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 0.7674684994272624,
|
| 960 |
+
"grad_norm": 8.430140495300293,
|
| 961 |
+
"learning_rate": 2.0357877092396775e-06,
|
| 962 |
+
"loss": 1.1568,
|
| 963 |
+
"step": 1340
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 0.7731958762886598,
|
| 967 |
+
"grad_norm": 10.301761627197266,
|
| 968 |
+
"learning_rate": 2.023161621494823e-06,
|
| 969 |
+
"loss": 1.1787,
|
| 970 |
+
"step": 1350
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.7789232531500573,
|
| 974 |
+
"grad_norm": 10.218490600585938,
|
| 975 |
+
"learning_rate": 2.0104931904965965e-06,
|
| 976 |
+
"loss": 1.0406,
|
| 977 |
+
"step": 1360
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 0.7846506300114547,
|
| 981 |
+
"grad_norm": 11.949338912963867,
|
| 982 |
+
"learning_rate": 1.9977834415927326e-06,
|
| 983 |
+
"loss": 1.1459,
|
| 984 |
+
"step": 1370
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.7903780068728522,
|
| 988 |
+
"grad_norm": 9.224528312683105,
|
| 989 |
+
"learning_rate": 1.985033403475123e-06,
|
| 990 |
+
"loss": 1.1283,
|
| 991 |
+
"step": 1380
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.7961053837342497,
|
| 995 |
+
"grad_norm": 9.380345344543457,
|
| 996 |
+
"learning_rate": 1.9722441080965565e-06,
|
| 997 |
+
"loss": 1.0648,
|
| 998 |
+
"step": 1390
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.8018327605956472,
|
| 1002 |
+
"grad_norm": 12.056816101074219,
|
| 1003 |
+
"learning_rate": 1.9594165905871968e-06,
|
| 1004 |
+
"loss": 1.1021,
|
| 1005 |
+
"step": 1400
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 0.8075601374570447,
|
| 1009 |
+
"grad_norm": 11.585986137390137,
|
| 1010 |
+
"learning_rate": 1.946551889170801e-06,
|
| 1011 |
+
"loss": 1.0863,
|
| 1012 |
+
"step": 1410
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 0.8132875143184422,
|
| 1016 |
+
"grad_norm": 10.934039115905762,
|
| 1017 |
+
"learning_rate": 1.9336510450806875e-06,
|
| 1018 |
+
"loss": 1.1676,
|
| 1019 |
+
"step": 1420
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 0.8190148911798396,
|
| 1023 |
+
"grad_norm": 8.694523811340332,
|
| 1024 |
+
"learning_rate": 1.920715102475464e-06,
|
| 1025 |
+
"loss": 1.033,
|
| 1026 |
+
"step": 1430
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 0.8247422680412371,
|
| 1030 |
+
"grad_norm": 9.625450134277344,
|
| 1031 |
+
"learning_rate": 1.9077451083545143e-06,
|
| 1032 |
+
"loss": 1.1275,
|
| 1033 |
+
"step": 1440
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 0.8304696449026346,
|
| 1037 |
+
"grad_norm": 7.567831516265869,
|
| 1038 |
+
"learning_rate": 1.8947421124732576e-06,
|
| 1039 |
+
"loss": 1.2054,
|
| 1040 |
+
"step": 1450
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 0.8361970217640321,
|
| 1044 |
+
"grad_norm": 10.554939270019531,
|
| 1045 |
+
"learning_rate": 1.8817071672581839e-06,
|
| 1046 |
+
"loss": 1.1276,
|
| 1047 |
+
"step": 1460
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 0.8419243986254296,
|
| 1051 |
+
"grad_norm": 10.357826232910156,
|
| 1052 |
+
"learning_rate": 1.868641327721675e-06,
|
| 1053 |
+
"loss": 1.2319,
|
| 1054 |
+
"step": 1470
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 0.847651775486827,
|
| 1058 |
+
"grad_norm": 16.142457962036133,
|
| 1059 |
+
"learning_rate": 1.8555456513766118e-06,
|
| 1060 |
+
"loss": 1.1066,
|
| 1061 |
+
"step": 1480
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 0.8533791523482245,
|
| 1065 |
+
"grad_norm": 9.302567481994629,
|
| 1066 |
+
"learning_rate": 1.8424211981507858e-06,
|
| 1067 |
+
"loss": 0.9964,
|
| 1068 |
+
"step": 1490
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 0.8591065292096219,
|
| 1072 |
+
"grad_norm": 13.468149185180664,
|
| 1073 |
+
"learning_rate": 1.8292690303011076e-06,
|
| 1074 |
+
"loss": 1.1009,
|
| 1075 |
+
"step": 1500
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 0.8591065292096219,
|
| 1079 |
+
"eval_loss": 1.086885690689087,
|
| 1080 |
+
"eval_runtime": 5.2843,
|
| 1081 |
+
"eval_samples_per_second": 68.127,
|
| 1082 |
+
"eval_steps_per_second": 17.032,
|
| 1083 |
+
"step": 1500
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 0.8648339060710195,
|
| 1087 |
+
"grad_norm": 9.433229446411133,
|
| 1088 |
+
"learning_rate": 1.816090212327634e-06,
|
| 1089 |
+
"loss": 1.0945,
|
| 1090 |
+
"step": 1510
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 0.870561282932417,
|
| 1094 |
+
"grad_norm": 12.731361389160156,
|
| 1095 |
+
"learning_rate": 1.8028858108874094e-06,
|
| 1096 |
+
"loss": 1.0626,
|
| 1097 |
+
"step": 1520
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 0.8762886597938144,
|
| 1101 |
+
"grad_norm": 10.284618377685547,
|
| 1102 |
+
"learning_rate": 1.7896568947081322e-06,
|
| 1103 |
+
"loss": 1.1405,
|
| 1104 |
+
"step": 1530
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 0.8820160366552119,
|
| 1108 |
+
"grad_norm": 11.09489631652832,
|
| 1109 |
+
"learning_rate": 1.776404534501656e-06,
|
| 1110 |
+
"loss": 1.1558,
|
| 1111 |
+
"step": 1540
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 0.8877434135166093,
|
| 1115 |
+
"grad_norm": 11.1835298538208,
|
| 1116 |
+
"learning_rate": 1.7631298028773296e-06,
|
| 1117 |
+
"loss": 1.2162,
|
| 1118 |
+
"step": 1550
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 0.8934707903780069,
|
| 1122 |
+
"grad_norm": 8.714800834655762,
|
| 1123 |
+
"learning_rate": 1.7498337742551817e-06,
|
| 1124 |
+
"loss": 1.0309,
|
| 1125 |
+
"step": 1560
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 0.8991981672394044,
|
| 1129 |
+
"grad_norm": 13.366388320922852,
|
| 1130 |
+
"learning_rate": 1.7365175247789623e-06,
|
| 1131 |
+
"loss": 1.0123,
|
| 1132 |
+
"step": 1570
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 0.9049255441008018,
|
| 1136 |
+
"grad_norm": 8.921965599060059,
|
| 1137 |
+
"learning_rate": 1.7231821322290399e-06,
|
| 1138 |
+
"loss": 1.1375,
|
| 1139 |
+
"step": 1580
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 0.9106529209621993,
|
| 1143 |
+
"grad_norm": 11.560654640197754,
|
| 1144 |
+
"learning_rate": 1.7098286759351717e-06,
|
| 1145 |
+
"loss": 1.1892,
|
| 1146 |
+
"step": 1590
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 0.9163802978235968,
|
| 1150 |
+
"grad_norm": 7.3394670486450195,
|
| 1151 |
+
"learning_rate": 1.6964582366891427e-06,
|
| 1152 |
+
"loss": 1.2526,
|
| 1153 |
+
"step": 1600
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 0.9221076746849943,
|
| 1157 |
+
"grad_norm": 8.516718864440918,
|
| 1158 |
+
"learning_rate": 1.6830718966572928e-06,
|
| 1159 |
+
"loss": 1.0736,
|
| 1160 |
+
"step": 1610
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 0.9278350515463918,
|
| 1164 |
+
"grad_norm": 10.884456634521484,
|
| 1165 |
+
"learning_rate": 1.6696707392929268e-06,
|
| 1166 |
+
"loss": 1.1471,
|
| 1167 |
+
"step": 1620
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 0.9335624284077892,
|
| 1171 |
+
"grad_norm": 11.866485595703125,
|
| 1172 |
+
"learning_rate": 1.6562558492486226e-06,
|
| 1173 |
+
"loss": 1.0814,
|
| 1174 |
+
"step": 1630
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 0.9392898052691867,
|
| 1178 |
+
"grad_norm": 10.002748489379883,
|
| 1179 |
+
"learning_rate": 1.6428283122884437e-06,
|
| 1180 |
+
"loss": 1.0393,
|
| 1181 |
+
"step": 1640
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 0.9450171821305842,
|
| 1185 |
+
"grad_norm": 10.888050079345703,
|
| 1186 |
+
"learning_rate": 1.6293892152000613e-06,
|
| 1187 |
+
"loss": 1.1509,
|
| 1188 |
+
"step": 1650
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 0.9507445589919816,
|
| 1192 |
+
"grad_norm": 10.640623092651367,
|
| 1193 |
+
"learning_rate": 1.6159396457067895e-06,
|
| 1194 |
+
"loss": 1.2364,
|
| 1195 |
+
"step": 1660
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 0.9564719358533792,
|
| 1199 |
+
"grad_norm": 10.190733909606934,
|
| 1200 |
+
"learning_rate": 1.60248069237955e-06,
|
| 1201 |
+
"loss": 1.203,
|
| 1202 |
+
"step": 1670
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 0.9621993127147767,
|
| 1206 |
+
"grad_norm": 9.43203353881836,
|
| 1207 |
+
"learning_rate": 1.5890134445487679e-06,
|
| 1208 |
+
"loss": 1.1801,
|
| 1209 |
+
"step": 1680
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 0.9679266895761741,
|
| 1213 |
+
"grad_norm": 11.540404319763184,
|
| 1214 |
+
"learning_rate": 1.5755389922162003e-06,
|
| 1215 |
+
"loss": 1.097,
|
| 1216 |
+
"step": 1690
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 0.9736540664375716,
|
| 1220 |
+
"grad_norm": 10.006051063537598,
|
| 1221 |
+
"learning_rate": 1.5620584259667175e-06,
|
| 1222 |
+
"loss": 1.1071,
|
| 1223 |
+
"step": 1700
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 0.979381443298969,
|
| 1227 |
+
"grad_norm": 11.146818161010742,
|
| 1228 |
+
"learning_rate": 1.548572836880034e-06,
|
| 1229 |
+
"loss": 1.297,
|
| 1230 |
+
"step": 1710
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 0.9851088201603666,
|
| 1234 |
+
"grad_norm": 15.32827091217041,
|
| 1235 |
+
"learning_rate": 1.5350833164423974e-06,
|
| 1236 |
+
"loss": 1.1533,
|
| 1237 |
+
"step": 1720
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"epoch": 0.9908361970217641,
|
| 1241 |
+
"grad_norm": 13.463982582092285,
|
| 1242 |
+
"learning_rate": 1.5215909564582499e-06,
|
| 1243 |
+
"loss": 1.1084,
|
| 1244 |
+
"step": 1730
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 0.9965635738831615,
|
| 1248 |
+
"grad_norm": 10.163274765014648,
|
| 1249 |
+
"learning_rate": 1.5080968489618567e-06,
|
| 1250 |
+
"loss": 1.0932,
|
| 1251 |
+
"step": 1740
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 1.002290950744559,
|
| 1255 |
+
"grad_norm": 8.775249481201172,
|
| 1256 |
+
"learning_rate": 1.494602086128923e-06,
|
| 1257 |
+
"loss": 0.9301,
|
| 1258 |
+
"step": 1750
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 1.0080183276059564,
|
| 1262 |
+
"grad_norm": 7.848943710327148,
|
| 1263 |
+
"learning_rate": 1.4811077601881953e-06,
|
| 1264 |
+
"loss": 0.6965,
|
| 1265 |
+
"step": 1760
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 1.013745704467354,
|
| 1269 |
+
"grad_norm": 13.538870811462402,
|
| 1270 |
+
"learning_rate": 1.4676149633330583e-06,
|
| 1271 |
+
"loss": 0.6828,
|
| 1272 |
+
"step": 1770
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 1.0194730813287514,
|
| 1276 |
+
"grad_norm": 10.99616527557373,
|
| 1277 |
+
"learning_rate": 1.4541247876331374e-06,
|
| 1278 |
+
"loss": 0.6853,
|
| 1279 |
+
"step": 1780
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 1.0252004581901488,
|
| 1283 |
+
"grad_norm": 8.934946060180664,
|
| 1284 |
+
"learning_rate": 1.440638324945909e-06,
|
| 1285 |
+
"loss": 0.759,
|
| 1286 |
+
"step": 1790
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 1.0309278350515463,
|
| 1290 |
+
"grad_norm": 8.413115501403809,
|
| 1291 |
+
"learning_rate": 1.4271566668283281e-06,
|
| 1292 |
+
"loss": 0.8167,
|
| 1293 |
+
"step": 1800
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 1.036655211912944,
|
| 1297 |
+
"grad_norm": 12.693419456481934,
|
| 1298 |
+
"learning_rate": 1.413680904448481e-06,
|
| 1299 |
+
"loss": 0.7765,
|
| 1300 |
+
"step": 1810
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 1.0423825887743414,
|
| 1304 |
+
"grad_norm": 18.639657974243164,
|
| 1305 |
+
"learning_rate": 1.4002121284972705e-06,
|
| 1306 |
+
"loss": 0.6784,
|
| 1307 |
+
"step": 1820
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 1.0481099656357389,
|
| 1311 |
+
"grad_norm": 9.834324836730957,
|
| 1312 |
+
"learning_rate": 1.3867514291001363e-06,
|
| 1313 |
+
"loss": 0.6933,
|
| 1314 |
+
"step": 1830
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 1.0538373424971363,
|
| 1318 |
+
"grad_norm": 10.812657356262207,
|
| 1319 |
+
"learning_rate": 1.373299895728824e-06,
|
| 1320 |
+
"loss": 0.7267,
|
| 1321 |
+
"step": 1840
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 1.0595647193585338,
|
| 1325 |
+
"grad_norm": 10.644402503967285,
|
| 1326 |
+
"learning_rate": 1.3598586171132068e-06,
|
| 1327 |
+
"loss": 0.7609,
|
| 1328 |
+
"step": 1850
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 1.0652920962199313,
|
| 1332 |
+
"grad_norm": 10.04351806640625,
|
| 1333 |
+
"learning_rate": 1.3464286811531663e-06,
|
| 1334 |
+
"loss": 0.6473,
|
| 1335 |
+
"step": 1860
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 1.0710194730813287,
|
| 1339 |
+
"grad_norm": 8.527778625488281,
|
| 1340 |
+
"learning_rate": 1.3330111748305409e-06,
|
| 1341 |
+
"loss": 0.6419,
|
| 1342 |
+
"step": 1870
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 1.0767468499427262,
|
| 1346 |
+
"grad_norm": 10.766478538513184,
|
| 1347 |
+
"learning_rate": 1.3196071841211486e-06,
|
| 1348 |
+
"loss": 0.707,
|
| 1349 |
+
"step": 1880
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 1.0824742268041236,
|
| 1353 |
+
"grad_norm": 9.123445510864258,
|
| 1354 |
+
"learning_rate": 1.3062177939068908e-06,
|
| 1355 |
+
"loss": 0.7146,
|
| 1356 |
+
"step": 1890
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 1.088201603665521,
|
| 1360 |
+
"grad_norm": 13.148359298706055,
|
| 1361 |
+
"learning_rate": 1.292844087887946e-06,
|
| 1362 |
+
"loss": 0.7023,
|
| 1363 |
+
"step": 1900
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 1.0939289805269188,
|
| 1367 |
+
"grad_norm": 9.834714889526367,
|
| 1368 |
+
"learning_rate": 1.279487148495056e-06,
|
| 1369 |
+
"loss": 0.6432,
|
| 1370 |
+
"step": 1910
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 1.0996563573883162,
|
| 1374 |
+
"grad_norm": 8.148127555847168,
|
| 1375 |
+
"learning_rate": 1.2661480568019203e-06,
|
| 1376 |
+
"loss": 0.6387,
|
| 1377 |
+
"step": 1920
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 1.1053837342497137,
|
| 1381 |
+
"grad_norm": 14.112127304077148,
|
| 1382 |
+
"learning_rate": 1.2528278924376933e-06,
|
| 1383 |
+
"loss": 0.7409,
|
| 1384 |
+
"step": 1930
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 1.1111111111111112,
|
| 1388 |
+
"grad_norm": 12.995244979858398,
|
| 1389 |
+
"learning_rate": 1.2395277334996047e-06,
|
| 1390 |
+
"loss": 0.7186,
|
| 1391 |
+
"step": 1940
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 1.1168384879725086,
|
| 1395 |
+
"grad_norm": 11.058789253234863,
|
| 1396 |
+
"learning_rate": 1.2262486564656995e-06,
|
| 1397 |
+
"loss": 0.7758,
|
| 1398 |
+
"step": 1950
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 1.122565864833906,
|
| 1402 |
+
"grad_norm": 11.76444149017334,
|
| 1403 |
+
"learning_rate": 1.2129917361077125e-06,
|
| 1404 |
+
"loss": 0.6512,
|
| 1405 |
+
"step": 1960
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 1.1282932416953035,
|
| 1409 |
+
"grad_norm": 12.445648193359375,
|
| 1410 |
+
"learning_rate": 1.1997580454040784e-06,
|
| 1411 |
+
"loss": 0.6401,
|
| 1412 |
+
"step": 1970
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 1.134020618556701,
|
| 1416 |
+
"grad_norm": 12.5171537399292,
|
| 1417 |
+
"learning_rate": 1.1865486554530874e-06,
|
| 1418 |
+
"loss": 0.7075,
|
| 1419 |
+
"step": 1980
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 1.1397479954180985,
|
| 1423 |
+
"grad_norm": 11.082114219665527,
|
| 1424 |
+
"learning_rate": 1.1733646353861951e-06,
|
| 1425 |
+
"loss": 0.7184,
|
| 1426 |
+
"step": 1990
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 1.145475372279496,
|
| 1430 |
+
"grad_norm": 8.820939064025879,
|
| 1431 |
+
"learning_rate": 1.1602070522814882e-06,
|
| 1432 |
+
"loss": 0.6901,
|
| 1433 |
+
"step": 2000
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 1.145475372279496,
|
| 1437 |
+
"eval_loss": 1.1485834121704102,
|
| 1438 |
+
"eval_runtime": 5.46,
|
| 1439 |
+
"eval_samples_per_second": 65.934,
|
| 1440 |
+
"eval_steps_per_second": 16.484,
|
| 1441 |
+
"step": 2000
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 1.1512027491408934,
|
| 1445 |
+
"grad_norm": 12.333765983581543,
|
| 1446 |
+
"learning_rate": 1.1470769710773189e-06,
|
| 1447 |
+
"loss": 0.7145,
|
| 1448 |
+
"step": 2010
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 1.1569301260022908,
|
| 1452 |
+
"grad_norm": 9.632255554199219,
|
| 1453 |
+
"learning_rate": 1.1339754544861111e-06,
|
| 1454 |
+
"loss": 0.6381,
|
| 1455 |
+
"step": 2020
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 1.1626575028636885,
|
| 1459 |
+
"grad_norm": 10.103231430053711,
|
| 1460 |
+
"learning_rate": 1.1209035629083498e-06,
|
| 1461 |
+
"loss": 0.6353,
|
| 1462 |
+
"step": 2030
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 1.168384879725086,
|
| 1466 |
+
"grad_norm": 11.000112533569336,
|
| 1467 |
+
"learning_rate": 1.1078623543467518e-06,
|
| 1468 |
+
"loss": 0.676,
|
| 1469 |
+
"step": 2040
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 1.1741122565864834,
|
| 1473 |
+
"grad_norm": 11.119033813476562,
|
| 1474 |
+
"learning_rate": 1.0948528843206357e-06,
|
| 1475 |
+
"loss": 0.6746,
|
| 1476 |
+
"step": 2050
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 1.179839633447881,
|
| 1480 |
+
"grad_norm": 11.437407493591309,
|
| 1481 |
+
"learning_rate": 1.081876205780492e-06,
|
| 1482 |
+
"loss": 0.6734,
|
| 1483 |
+
"step": 2060
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 1.1855670103092784,
|
| 1487 |
+
"grad_norm": 8.372086524963379,
|
| 1488 |
+
"learning_rate": 1.068933369022758e-06,
|
| 1489 |
+
"loss": 0.7758,
|
| 1490 |
+
"step": 2070
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 1.1912943871706758,
|
| 1494 |
+
"grad_norm": 7.501805305480957,
|
| 1495 |
+
"learning_rate": 1.0560254216048105e-06,
|
| 1496 |
+
"loss": 0.8655,
|
| 1497 |
+
"step": 2080
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 1.1970217640320733,
|
| 1501 |
+
"grad_norm": 11.55257797241211,
|
| 1502 |
+
"learning_rate": 1.0431534082601814e-06,
|
| 1503 |
+
"loss": 0.7701,
|
| 1504 |
+
"step": 2090
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 1.2027491408934707,
|
| 1508 |
+
"grad_norm": 13.704492568969727,
|
| 1509 |
+
"learning_rate": 1.0303183708139966e-06,
|
| 1510 |
+
"loss": 0.7325,
|
| 1511 |
+
"step": 2100
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 1.2084765177548682,
|
| 1515 |
+
"grad_norm": 9.895234107971191,
|
| 1516 |
+
"learning_rate": 1.0175213480986553e-06,
|
| 1517 |
+
"loss": 0.7678,
|
| 1518 |
+
"step": 2110
|
| 1519 |
+
},
|
| 1520 |
+
{
|
| 1521 |
+
"epoch": 1.2142038946162657,
|
| 1522 |
+
"grad_norm": 11.093770027160645,
|
| 1523 |
+
"learning_rate": 1.0047633758697507e-06,
|
| 1524 |
+
"loss": 0.7074,
|
| 1525 |
+
"step": 2120
|
| 1526 |
+
},
|
| 1527 |
+
{
|
| 1528 |
+
"epoch": 1.2199312714776633,
|
| 1529 |
+
"grad_norm": 7.835501194000244,
|
| 1530 |
+
"learning_rate": 9.920454867222362e-07,
|
| 1531 |
+
"loss": 0.6939,
|
| 1532 |
+
"step": 2130
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 1.2256586483390608,
|
| 1536 |
+
"grad_norm": 12.765395164489746,
|
| 1537 |
+
"learning_rate": 9.793687100068537e-07,
|
| 1538 |
+
"loss": 0.6601,
|
| 1539 |
+
"step": 2140
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 1.2313860252004583,
|
| 1543 |
+
"grad_norm": 8.465944290161133,
|
| 1544 |
+
"learning_rate": 9.667340717468165e-07,
|
| 1545 |
+
"loss": 0.6601,
|
| 1546 |
+
"step": 2150
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 1.2371134020618557,
|
| 1550 |
+
"grad_norm": 8.568673133850098,
|
| 1551 |
+
"learning_rate": 9.54142594554769e-07,
|
| 1552 |
+
"loss": 0.6536,
|
| 1553 |
+
"step": 2160
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 1.2428407789232532,
|
| 1557 |
+
"grad_norm": 11.03416633605957,
|
| 1558 |
+
"learning_rate": 9.415952975500177e-07,
|
| 1559 |
+
"loss": 0.7267,
|
| 1560 |
+
"step": 2170
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 1.2485681557846506,
|
| 1564 |
+
"grad_norm": 12.343921661376953,
|
| 1565 |
+
"learning_rate": 9.290931962760477e-07,
|
| 1566 |
+
"loss": 0.7921,
|
| 1567 |
+
"step": 2180
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"epoch": 1.254295532646048,
|
| 1571 |
+
"grad_norm": 12.319411277770996,
|
| 1572 |
+
"learning_rate": 9.166373026183269e-07,
|
| 1573 |
+
"loss": 0.6843,
|
| 1574 |
+
"step": 2190
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 1.2600229095074456,
|
| 1578 |
+
"grad_norm": 11.100428581237793,
|
| 1579 |
+
"learning_rate": 9.042286247224054e-07,
|
| 1580 |
+
"loss": 0.7535,
|
| 1581 |
+
"step": 2200
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 1.265750286368843,
|
| 1585 |
+
"grad_norm": 10.030117988586426,
|
| 1586 |
+
"learning_rate": 8.918681669123218e-07,
|
| 1587 |
+
"loss": 0.6571,
|
| 1588 |
+
"step": 2210
|
| 1589 |
+
},
|
| 1590 |
+
{
|
| 1591 |
+
"epoch": 1.2714776632302405,
|
| 1592 |
+
"grad_norm": 10.269767761230469,
|
| 1593 |
+
"learning_rate": 8.795569296093133e-07,
|
| 1594 |
+
"loss": 0.7255,
|
| 1595 |
+
"step": 2220
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 1.277205040091638,
|
| 1599 |
+
"grad_norm": 11.503005027770996,
|
| 1600 |
+
"learning_rate": 8.67295909250846e-07,
|
| 1601 |
+
"loss": 0.7021,
|
| 1602 |
+
"step": 2230
|
| 1603 |
+
},
|
| 1604 |
+
{
|
| 1605 |
+
"epoch": 1.2829324169530354,
|
| 1606 |
+
"grad_norm": 19.0956974029541,
|
| 1607 |
+
"learning_rate": 8.55086098209965e-07,
|
| 1608 |
+
"loss": 0.644,
|
| 1609 |
+
"step": 2240
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 1.2886597938144329,
|
| 1613 |
+
"grad_norm": 11.60425090789795,
|
| 1614 |
+
"learning_rate": 8.429284847149765e-07,
|
| 1615 |
+
"loss": 0.7289,
|
| 1616 |
+
"step": 2250
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 1.2943871706758305,
|
| 1620 |
+
"grad_norm": 11.15966510772705,
|
| 1621 |
+
"learning_rate": 8.308240527694602e-07,
|
| 1622 |
+
"loss": 0.6789,
|
| 1623 |
+
"step": 2260
|
| 1624 |
+
},
|
| 1625 |
+
{
|
| 1626 |
+
"epoch": 1.300114547537228,
|
| 1627 |
+
"grad_norm": 11.775574684143066,
|
| 1628 |
+
"learning_rate": 8.187737820726293e-07,
|
| 1629 |
+
"loss": 0.7226,
|
| 1630 |
+
"step": 2270
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 1.3058419243986255,
|
| 1634 |
+
"grad_norm": 10.303918838500977,
|
| 1635 |
+
"learning_rate": 8.067786479400346e-07,
|
| 1636 |
+
"loss": 0.6405,
|
| 1637 |
+
"step": 2280
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 1.311569301260023,
|
| 1641 |
+
"grad_norm": 11.979586601257324,
|
| 1642 |
+
"learning_rate": 7.948396212246273e-07,
|
| 1643 |
+
"loss": 0.6847,
|
| 1644 |
+
"step": 2290
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 1.3172966781214204,
|
| 1648 |
+
"grad_norm": 8.767427444458008,
|
| 1649 |
+
"learning_rate": 7.829576682381782e-07,
|
| 1650 |
+
"loss": 0.681,
|
| 1651 |
+
"step": 2300
|
| 1652 |
+
},
|
| 1653 |
+
{
|
| 1654 |
+
"epoch": 1.3230240549828178,
|
| 1655 |
+
"grad_norm": 12.378206253051758,
|
| 1656 |
+
"learning_rate": 7.711337506730695e-07,
|
| 1657 |
+
"loss": 0.7108,
|
| 1658 |
+
"step": 2310
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 1.3287514318442153,
|
| 1662 |
+
"grad_norm": 9.478968620300293,
|
| 1663 |
+
"learning_rate": 7.593688255244566e-07,
|
| 1664 |
+
"loss": 0.6728,
|
| 1665 |
+
"step": 2320
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 1.3344788087056128,
|
| 1669 |
+
"grad_norm": 13.136493682861328,
|
| 1670 |
+
"learning_rate": 7.476638450128123e-07,
|
| 1671 |
+
"loss": 0.6783,
|
| 1672 |
+
"step": 2330
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 1.3402061855670104,
|
| 1676 |
+
"grad_norm": 12.096550941467285,
|
| 1677 |
+
"learning_rate": 7.360197565068561e-07,
|
| 1678 |
+
"loss": 0.6398,
|
| 1679 |
+
"step": 2340
|
| 1680 |
+
},
|
| 1681 |
+
{
|
| 1682 |
+
"epoch": 1.345933562428408,
|
| 1683 |
+
"grad_norm": 9.796936988830566,
|
| 1684 |
+
"learning_rate": 7.244375024468765e-07,
|
| 1685 |
+
"loss": 0.659,
|
| 1686 |
+
"step": 2350
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 1.3516609392898054,
|
| 1690 |
+
"grad_norm": 12.897764205932617,
|
| 1691 |
+
"learning_rate": 7.129180202684554e-07,
|
| 1692 |
+
"loss": 0.6226,
|
| 1693 |
+
"step": 2360
|
| 1694 |
+
},
|
| 1695 |
+
{
|
| 1696 |
+
"epoch": 1.3573883161512028,
|
| 1697 |
+
"grad_norm": 13.480414390563965,
|
| 1698 |
+
"learning_rate": 7.0146224232659e-07,
|
| 1699 |
+
"loss": 0.7311,
|
| 1700 |
+
"step": 2370
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 1.3631156930126003,
|
| 1704 |
+
"grad_norm": 9.178197860717773,
|
| 1705 |
+
"learning_rate": 6.900710958202341e-07,
|
| 1706 |
+
"loss": 0.7167,
|
| 1707 |
+
"step": 2380
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 1.3688430698739977,
|
| 1711 |
+
"grad_norm": 7.676974773406982,
|
| 1712 |
+
"learning_rate": 6.787455027172515e-07,
|
| 1713 |
+
"loss": 0.7085,
|
| 1714 |
+
"step": 2390
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 1.3745704467353952,
|
| 1718 |
+
"grad_norm": 19.0358829498291,
|
| 1719 |
+
"learning_rate": 6.674863796797954e-07,
|
| 1720 |
+
"loss": 0.6221,
|
| 1721 |
+
"step": 2400
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 1.3802978235967927,
|
| 1725 |
+
"grad_norm": 10.022867202758789,
|
| 1726 |
+
"learning_rate": 6.562946379901158e-07,
|
| 1727 |
+
"loss": 0.7337,
|
| 1728 |
+
"step": 2410
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 1.38602520045819,
|
| 1732 |
+
"grad_norm": 11.5038480758667,
|
| 1733 |
+
"learning_rate": 6.451711834768026e-07,
|
| 1734 |
+
"loss": 0.6849,
|
| 1735 |
+
"step": 2420
|
| 1736 |
+
},
|
| 1737 |
+
{
|
| 1738 |
+
"epoch": 1.3917525773195876,
|
| 1739 |
+
"grad_norm": 10.91701889038086,
|
| 1740 |
+
"learning_rate": 6.341169164414712e-07,
|
| 1741 |
+
"loss": 0.6061,
|
| 1742 |
+
"step": 2430
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 1.397479954180985,
|
| 1746 |
+
"grad_norm": 11.305731773376465,
|
| 1747 |
+
"learning_rate": 6.23132731585894e-07,
|
| 1748 |
+
"loss": 0.6072,
|
| 1749 |
+
"step": 2440
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 1.4032073310423825,
|
| 1753 |
+
"grad_norm": 11.2603178024292,
|
| 1754 |
+
"learning_rate": 6.122195179395854e-07,
|
| 1755 |
+
"loss": 0.7589,
|
| 1756 |
+
"step": 2450
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 1.40893470790378,
|
| 1760 |
+
"grad_norm": 10.891215324401855,
|
| 1761 |
+
"learning_rate": 6.013781587878464e-07,
|
| 1762 |
+
"loss": 0.6253,
|
| 1763 |
+
"step": 2460
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 1.4146620847651776,
|
| 1767 |
+
"grad_norm": 13.905660629272461,
|
| 1768 |
+
"learning_rate": 5.906095316002755e-07,
|
| 1769 |
+
"loss": 0.7618,
|
| 1770 |
+
"step": 2470
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 1.420389461626575,
|
| 1774 |
+
"grad_norm": 8.74496078491211,
|
| 1775 |
+
"learning_rate": 5.799145079597458e-07,
|
| 1776 |
+
"loss": 0.6866,
|
| 1777 |
+
"step": 2480
|
| 1778 |
+
},
|
| 1779 |
+
{
|
| 1780 |
+
"epoch": 1.4261168384879725,
|
| 1781 |
+
"grad_norm": 10.08084487915039,
|
| 1782 |
+
"learning_rate": 5.69293953491864e-07,
|
| 1783 |
+
"loss": 0.6584,
|
| 1784 |
+
"step": 2490
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 1.43184421534937,
|
| 1788 |
+
"grad_norm": 9.269981384277344,
|
| 1789 |
+
"learning_rate": 5.587487277949075e-07,
|
| 1790 |
+
"loss": 0.7377,
|
| 1791 |
+
"step": 2500
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"epoch": 1.43184421534937,
|
| 1795 |
+
"eval_loss": 1.1487523317337036,
|
| 1796 |
+
"eval_runtime": 5.3152,
|
| 1797 |
+
"eval_samples_per_second": 67.73,
|
| 1798 |
+
"eval_steps_per_second": 16.933,
|
| 1799 |
+
"step": 2500
|
| 1800 |
+
},
|
| 1801 |
+
{
|
| 1802 |
+
"epoch": 1.4375715922107675,
|
| 1803 |
+
"grad_norm": 8.877992630004883,
|
| 1804 |
+
"learning_rate": 5.48279684370252e-07,
|
| 1805 |
+
"loss": 0.7111,
|
| 1806 |
+
"step": 2510
|
| 1807 |
+
},
|
| 1808 |
+
{
|
| 1809 |
+
"epoch": 1.443298969072165,
|
| 1810 |
+
"grad_norm": 24.23348045349121,
|
| 1811 |
+
"learning_rate": 5.378876705532904e-07,
|
| 1812 |
+
"loss": 0.62,
|
| 1813 |
+
"step": 2520
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 1.4490263459335624,
|
| 1817 |
+
"grad_norm": 9.67536735534668,
|
| 1818 |
+
"learning_rate": 5.275735274448529e-07,
|
| 1819 |
+
"loss": 0.7702,
|
| 1820 |
+
"step": 2530
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 1.4547537227949598,
|
| 1824 |
+
"grad_norm": 10.30777359008789,
|
| 1825 |
+
"learning_rate": 5.173380898431305e-07,
|
| 1826 |
+
"loss": 0.6682,
|
| 1827 |
+
"step": 2540
|
| 1828 |
+
},
|
| 1829 |
+
{
|
| 1830 |
+
"epoch": 1.4604810996563573,
|
| 1831 |
+
"grad_norm": 12.1168212890625,
|
| 1832 |
+
"learning_rate": 5.071821861761074e-07,
|
| 1833 |
+
"loss": 0.6713,
|
| 1834 |
+
"step": 2550
|
| 1835 |
+
},
|
| 1836 |
+
{
|
| 1837 |
+
"epoch": 1.466208476517755,
|
| 1838 |
+
"grad_norm": 7.113183975219727,
|
| 1839 |
+
"learning_rate": 4.97106638434512e-07,
|
| 1840 |
+
"loss": 0.7582,
|
| 1841 |
+
"step": 2560
|
| 1842 |
+
},
|
| 1843 |
+
{
|
| 1844 |
+
"epoch": 1.4719358533791524,
|
| 1845 |
+
"grad_norm": 13.592216491699219,
|
| 1846 |
+
"learning_rate": 4.87112262105286e-07,
|
| 1847 |
+
"loss": 0.6917,
|
| 1848 |
+
"step": 2570
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"epoch": 1.47766323024055,
|
| 1852 |
+
"grad_norm": 10.175348281860352,
|
| 1853 |
+
"learning_rate": 4.771998661055823e-07,
|
| 1854 |
+
"loss": 0.7341,
|
| 1855 |
+
"step": 2580
|
| 1856 |
+
},
|
| 1857 |
+
{
|
| 1858 |
+
"epoch": 1.4833906071019474,
|
| 1859 |
+
"grad_norm": 10.051101684570312,
|
| 1860 |
+
"learning_rate": 4.673702527172919e-07,
|
| 1861 |
+
"loss": 0.5795,
|
| 1862 |
+
"step": 2590
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 1.4891179839633448,
|
| 1866 |
+
"grad_norm": 9.961878776550293,
|
| 1867 |
+
"learning_rate": 4.576242175221121e-07,
|
| 1868 |
+
"loss": 0.6895,
|
| 1869 |
+
"step": 2600
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 1.4948453608247423,
|
| 1873 |
+
"grad_norm": 10.214471817016602,
|
| 1874 |
+
"learning_rate": 4.479625493371512e-07,
|
| 1875 |
+
"loss": 0.6552,
|
| 1876 |
+
"step": 2610
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 1.5005727376861397,
|
| 1880 |
+
"grad_norm": 16.537036895751953,
|
| 1881 |
+
"learning_rate": 4.3838603015108537e-07,
|
| 1882 |
+
"loss": 0.6821,
|
| 1883 |
+
"step": 2620
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 1.5063001145475372,
|
| 1887 |
+
"grad_norm": 8.615083694458008,
|
| 1888 |
+
"learning_rate": 4.288954350608672e-07,
|
| 1889 |
+
"loss": 0.6581,
|
| 1890 |
+
"step": 2630
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"epoch": 1.5120274914089347,
|
| 1894 |
+
"grad_norm": 12.914610862731934,
|
| 1895 |
+
"learning_rate": 4.1949153220898987e-07,
|
| 1896 |
+
"loss": 0.7688,
|
| 1897 |
+
"step": 2640
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 1.5177548682703321,
|
| 1901 |
+
"grad_norm": 10.495301246643066,
|
| 1902 |
+
"learning_rate": 4.1017508272131794e-07,
|
| 1903 |
+
"loss": 0.6831,
|
| 1904 |
+
"step": 2650
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 1.5234822451317296,
|
| 1908 |
+
"grad_norm": 11.30919075012207,
|
| 1909 |
+
"learning_rate": 4.009468406454818e-07,
|
| 1910 |
+
"loss": 0.796,
|
| 1911 |
+
"step": 2660
|
| 1912 |
+
},
|
| 1913 |
+
{
|
| 1914 |
+
"epoch": 1.529209621993127,
|
| 1915 |
+
"grad_norm": 9.941025733947754,
|
| 1916 |
+
"learning_rate": 3.9180755288984837e-07,
|
| 1917 |
+
"loss": 0.7486,
|
| 1918 |
+
"step": 2670
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"epoch": 1.5349369988545245,
|
| 1922 |
+
"grad_norm": 11.294404983520508,
|
| 1923 |
+
"learning_rate": 3.8275795916306827e-07,
|
| 1924 |
+
"loss": 0.6736,
|
| 1925 |
+
"step": 2680
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 1.540664375715922,
|
| 1929 |
+
"grad_norm": 12.294686317443848,
|
| 1930 |
+
"learning_rate": 3.737987919142061e-07,
|
| 1931 |
+
"loss": 0.6431,
|
| 1932 |
+
"step": 2690
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"epoch": 1.5463917525773194,
|
| 1936 |
+
"grad_norm": 11.265865325927734,
|
| 1937 |
+
"learning_rate": 3.64930776273457e-07,
|
| 1938 |
+
"loss": 0.8467,
|
| 1939 |
+
"step": 2700
|
| 1940 |
+
},
|
| 1941 |
+
{
|
| 1942 |
+
"epoch": 1.552119129438717,
|
| 1943 |
+
"grad_norm": 7.701663970947266,
|
| 1944 |
+
"learning_rate": 3.561546299934577e-07,
|
| 1945 |
+
"loss": 0.7307,
|
| 1946 |
+
"step": 2710
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 1.5578465063001146,
|
| 1950 |
+
"grad_norm": 14.973838806152344,
|
| 1951 |
+
"learning_rate": 3.4747106339119276e-07,
|
| 1952 |
+
"loss": 0.7451,
|
| 1953 |
+
"step": 2720
|
| 1954 |
+
},
|
| 1955 |
+
{
|
| 1956 |
+
"epoch": 1.563573883161512,
|
| 1957 |
+
"grad_norm": 23.487585067749023,
|
| 1958 |
+
"learning_rate": 3.388807792905045e-07,
|
| 1959 |
+
"loss": 0.6253,
|
| 1960 |
+
"step": 2730
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 1.5693012600229095,
|
| 1964 |
+
"grad_norm": 10.245492935180664,
|
| 1965 |
+
"learning_rate": 3.303844729652073e-07,
|
| 1966 |
+
"loss": 0.7833,
|
| 1967 |
+
"step": 2740
|
| 1968 |
+
},
|
| 1969 |
+
{
|
| 1970 |
+
"epoch": 1.575028636884307,
|
| 1971 |
+
"grad_norm": 10.44719123840332,
|
| 1972 |
+
"learning_rate": 3.219828320828148e-07,
|
| 1973 |
+
"loss": 0.6858,
|
| 1974 |
+
"step": 2750
|
| 1975 |
+
},
|
| 1976 |
+
{
|
| 1977 |
+
"epoch": 1.5807560137457046,
|
| 1978 |
+
"grad_norm": 8.665092468261719,
|
| 1979 |
+
"learning_rate": 3.1367653664888173e-07,
|
| 1980 |
+
"loss": 0.6299,
|
| 1981 |
+
"step": 2760
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"epoch": 1.586483390607102,
|
| 1985 |
+
"grad_norm": 11.613511085510254,
|
| 1986 |
+
"learning_rate": 3.054662589519661e-07,
|
| 1987 |
+
"loss": 0.8032,
|
| 1988 |
+
"step": 2770
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 1.5922107674684995,
|
| 1992 |
+
"grad_norm": 7.786969184875488,
|
| 1993 |
+
"learning_rate": 2.973526635092163e-07,
|
| 1994 |
+
"loss": 0.7187,
|
| 1995 |
+
"step": 2780
|
| 1996 |
+
},
|
| 1997 |
+
{
|
| 1998 |
+
"epoch": 1.597938144329897,
|
| 1999 |
+
"grad_norm": 13.719048500061035,
|
| 2000 |
+
"learning_rate": 2.8933640701258597e-07,
|
| 2001 |
+
"loss": 0.7349,
|
| 2002 |
+
"step": 2790
|
| 2003 |
+
},
|
| 2004 |
+
{
|
| 2005 |
+
"epoch": 1.6036655211912945,
|
| 2006 |
+
"grad_norm": 11.509296417236328,
|
| 2007 |
+
"learning_rate": 2.8141813827568525e-07,
|
| 2008 |
+
"loss": 0.6635,
|
| 2009 |
+
"step": 2800
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 1.609392898052692,
|
| 2013 |
+
"grad_norm": 10.817230224609375,
|
| 2014 |
+
"learning_rate": 2.7359849818126533e-07,
|
| 2015 |
+
"loss": 0.7161,
|
| 2016 |
+
"step": 2810
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 1.6151202749140894,
|
| 2020 |
+
"grad_norm": 13.700976371765137,
|
| 2021 |
+
"learning_rate": 2.6587811962934823e-07,
|
| 2022 |
+
"loss": 0.7205,
|
| 2023 |
+
"step": 2820
|
| 2024 |
+
},
|
| 2025 |
+
{
|
| 2026 |
+
"epoch": 1.6208476517754868,
|
| 2027 |
+
"grad_norm": 10.548551559448242,
|
| 2028 |
+
"learning_rate": 2.582576274860015e-07,
|
| 2029 |
+
"loss": 0.6782,
|
| 2030 |
+
"step": 2830
|
| 2031 |
+
},
|
| 2032 |
+
{
|
| 2033 |
+
"epoch": 1.6265750286368843,
|
| 2034 |
+
"grad_norm": 9.971668243408203,
|
| 2035 |
+
"learning_rate": 2.507376385327634e-07,
|
| 2036 |
+
"loss": 0.7375,
|
| 2037 |
+
"step": 2840
|
| 2038 |
+
},
|
| 2039 |
+
{
|
| 2040 |
+
"epoch": 1.6323024054982818,
|
| 2041 |
+
"grad_norm": 11.734025001525879,
|
| 2042 |
+
"learning_rate": 2.4331876141672204e-07,
|
| 2043 |
+
"loss": 0.6543,
|
| 2044 |
+
"step": 2850
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 1.6380297823596792,
|
| 2048 |
+
"grad_norm": 10.10330581665039,
|
| 2049 |
+
"learning_rate": 2.360015966012537e-07,
|
| 2050 |
+
"loss": 0.7008,
|
| 2051 |
+
"step": 2860
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"epoch": 1.6437571592210767,
|
| 2055 |
+
"grad_norm": 8.163260459899902,
|
| 2056 |
+
"learning_rate": 2.2878673631742236e-07,
|
| 2057 |
+
"loss": 0.758,
|
| 2058 |
+
"step": 2870
|
| 2059 |
+
},
|
| 2060 |
+
{
|
| 2061 |
+
"epoch": 1.6494845360824741,
|
| 2062 |
+
"grad_norm": 9.96818733215332,
|
| 2063 |
+
"learning_rate": 2.2167476451604624e-07,
|
| 2064 |
+
"loss": 0.7625,
|
| 2065 |
+
"step": 2880
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 1.6552119129438716,
|
| 2069 |
+
"grad_norm": 11.303291320800781,
|
| 2070 |
+
"learning_rate": 2.1466625682043516e-07,
|
| 2071 |
+
"loss": 0.6977,
|
| 2072 |
+
"step": 2890
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 1.660939289805269,
|
| 2076 |
+
"grad_norm": 11.4495210647583,
|
| 2077 |
+
"learning_rate": 2.0776178047979978e-07,
|
| 2078 |
+
"loss": 0.7007,
|
| 2079 |
+
"step": 2900
|
| 2080 |
+
},
|
| 2081 |
+
{
|
| 2082 |
+
"epoch": 1.6666666666666665,
|
| 2083 |
+
"grad_norm": 13.028532981872559,
|
| 2084 |
+
"learning_rate": 2.0096189432334195e-07,
|
| 2085 |
+
"loss": 0.6004,
|
| 2086 |
+
"step": 2910
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"epoch": 1.6723940435280642,
|
| 2090 |
+
"grad_norm": 10.361539840698242,
|
| 2091 |
+
"learning_rate": 1.9426714871502292e-07,
|
| 2092 |
+
"loss": 0.7078,
|
| 2093 |
+
"step": 2920
|
| 2094 |
+
},
|
| 2095 |
+
{
|
| 2096 |
+
"epoch": 1.6781214203894617,
|
| 2097 |
+
"grad_norm": 10.60468864440918,
|
| 2098 |
+
"learning_rate": 1.8767808550901911e-07,
|
| 2099 |
+
"loss": 0.7244,
|
| 2100 |
+
"step": 2930
|
| 2101 |
+
},
|
| 2102 |
+
{
|
| 2103 |
+
"epoch": 1.6838487972508591,
|
| 2104 |
+
"grad_norm": 9.203190803527832,
|
| 2105 |
+
"learning_rate": 1.811952380058657e-07,
|
| 2106 |
+
"loss": 0.636,
|
| 2107 |
+
"step": 2940
|
| 2108 |
+
},
|
| 2109 |
+
{
|
| 2110 |
+
"epoch": 1.6895761741122566,
|
| 2111 |
+
"grad_norm": 11.939068794250488,
|
| 2112 |
+
"learning_rate": 1.7481913090929314e-07,
|
| 2113 |
+
"loss": 0.7098,
|
| 2114 |
+
"step": 2950
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"epoch": 1.695303550973654,
|
| 2118 |
+
"grad_norm": 9.831012725830078,
|
| 2119 |
+
"learning_rate": 1.6855028028375897e-07,
|
| 2120 |
+
"loss": 0.6872,
|
| 2121 |
+
"step": 2960
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 1.7010309278350515,
|
| 2125 |
+
"grad_norm": 18.38079261779785,
|
| 2126 |
+
"learning_rate": 1.623891935126786e-07,
|
| 2127 |
+
"loss": 0.6811,
|
| 2128 |
+
"step": 2970
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 1.7067583046964492,
|
| 2132 |
+
"grad_norm": 9.422789573669434,
|
| 2133 |
+
"learning_rate": 1.5633636925735933e-07,
|
| 2134 |
+
"loss": 0.6251,
|
| 2135 |
+
"step": 2980
|
| 2136 |
+
},
|
| 2137 |
+
{
|
| 2138 |
+
"epoch": 1.7124856815578466,
|
| 2139 |
+
"grad_norm": 9.562675476074219,
|
| 2140 |
+
"learning_rate": 1.503922974166406e-07,
|
| 2141 |
+
"loss": 0.7189,
|
| 2142 |
+
"step": 2990
|
| 2143 |
+
},
|
| 2144 |
+
{
|
| 2145 |
+
"epoch": 1.718213058419244,
|
| 2146 |
+
"grad_norm": 9.7070894241333,
|
| 2147 |
+
"learning_rate": 1.4455745908724226e-07,
|
| 2148 |
+
"loss": 0.6947,
|
| 2149 |
+
"step": 3000
|
| 2150 |
+
},
|
| 2151 |
+
{
|
| 2152 |
+
"epoch": 1.718213058419244,
|
| 2153 |
+
"eval_loss": 1.150073528289795,
|
| 2154 |
+
"eval_runtime": 5.3064,
|
| 2155 |
+
"eval_samples_per_second": 67.842,
|
| 2156 |
+
"eval_steps_per_second": 16.961,
|
| 2157 |
+
"step": 3000
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 1.7239404352806416,
|
| 2161 |
+
"grad_norm": 12.27785873413086,
|
| 2162 |
+
"learning_rate": 1.3883232652482553e-07,
|
| 2163 |
+
"loss": 0.8112,
|
| 2164 |
+
"step": 3010
|
| 2165 |
+
},
|
| 2166 |
+
{
|
| 2167 |
+
"epoch": 1.729667812142039,
|
| 2168 |
+
"grad_norm": 12.872515678405762,
|
| 2169 |
+
"learning_rate": 1.3321736310577164e-07,
|
| 2170 |
+
"loss": 0.7575,
|
| 2171 |
+
"step": 3020
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"epoch": 1.7353951890034365,
|
| 2175 |
+
"grad_norm": 11.842426300048828,
|
| 2176 |
+
"learning_rate": 1.2771302328967577e-07,
|
| 2177 |
+
"loss": 0.7448,
|
| 2178 |
+
"step": 3030
|
| 2179 |
+
},
|
| 2180 |
+
{
|
| 2181 |
+
"epoch": 1.741122565864834,
|
| 2182 |
+
"grad_norm": 11.257376670837402,
|
| 2183 |
+
"learning_rate": 1.2231975258256462e-07,
|
| 2184 |
+
"loss": 0.6947,
|
| 2185 |
+
"step": 3040
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 1.7468499427262314,
|
| 2189 |
+
"grad_norm": 12.237911224365234,
|
| 2190 |
+
"learning_rate": 1.1703798750083938e-07,
|
| 2191 |
+
"loss": 0.684,
|
| 2192 |
+
"step": 3050
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 1.7525773195876289,
|
| 2196 |
+
"grad_norm": 11.303350448608398,
|
| 2197 |
+
"learning_rate": 1.1186815553594382e-07,
|
| 2198 |
+
"loss": 0.7047,
|
| 2199 |
+
"step": 3060
|
| 2200 |
+
},
|
| 2201 |
+
{
|
| 2202 |
+
"epoch": 1.7583046964490263,
|
| 2203 |
+
"grad_norm": 10.056689262390137,
|
| 2204 |
+
"learning_rate": 1.068106751197659e-07,
|
| 2205 |
+
"loss": 0.6862,
|
| 2206 |
+
"step": 3070
|
| 2207 |
+
},
|
| 2208 |
+
{
|
| 2209 |
+
"epoch": 1.7640320733104238,
|
| 2210 |
+
"grad_norm": 9.765776634216309,
|
| 2211 |
+
"learning_rate": 1.0186595559076939e-07,
|
| 2212 |
+
"loss": 0.7131,
|
| 2213 |
+
"step": 3080
|
| 2214 |
+
},
|
| 2215 |
+
{
|
| 2216 |
+
"epoch": 1.7697594501718212,
|
| 2217 |
+
"grad_norm": 10.987778663635254,
|
| 2218 |
+
"learning_rate": 9.703439716086493e-08,
|
| 2219 |
+
"loss": 0.6012,
|
| 2220 |
+
"step": 3090
|
| 2221 |
+
},
|
| 2222 |
+
{
|
| 2223 |
+
"epoch": 1.7754868270332187,
|
| 2224 |
+
"grad_norm": 14.675382614135742,
|
| 2225 |
+
"learning_rate": 9.231639088301635e-08,
|
| 2226 |
+
"loss": 0.6843,
|
| 2227 |
+
"step": 3100
|
| 2228 |
+
},
|
| 2229 |
+
{
|
| 2230 |
+
"epoch": 1.7812142038946162,
|
| 2231 |
+
"grad_norm": 15.27701187133789,
|
| 2232 |
+
"learning_rate": 8.771231861959106e-08,
|
| 2233 |
+
"loss": 0.7452,
|
| 2234 |
+
"step": 3110
|
| 2235 |
+
},
|
| 2236 |
+
{
|
| 2237 |
+
"epoch": 1.7869415807560136,
|
| 2238 |
+
"grad_norm": 12.017073631286621,
|
| 2239 |
+
"learning_rate": 8.322255301145204e-08,
|
| 2240 |
+
"loss": 0.7111,
|
| 2241 |
+
"step": 3120
|
| 2242 |
+
},
|
| 2243 |
+
{
|
| 2244 |
+
"epoch": 1.792668957617411,
|
| 2245 |
+
"grad_norm": 8.804431915283203,
|
| 2246 |
+
"learning_rate": 7.884745744779937e-08,
|
| 2247 |
+
"loss": 0.7718,
|
| 2248 |
+
"step": 3130
|
| 2249 |
+
},
|
| 2250 |
+
{
|
| 2251 |
+
"epoch": 1.7983963344788088,
|
| 2252 |
+
"grad_norm": 10.447242736816406,
|
| 2253 |
+
"learning_rate": 7.458738603675602e-08,
|
| 2254 |
+
"loss": 0.8338,
|
| 2255 |
+
"step": 3140
|
| 2256 |
+
},
|
| 2257 |
+
{
|
| 2258 |
+
"epoch": 1.8041237113402062,
|
| 2259 |
+
"grad_norm": 13.058491706848145,
|
| 2260 |
+
"learning_rate": 7.044268357670869e-08,
|
| 2261 |
+
"loss": 0.5838,
|
| 2262 |
+
"step": 3150
|
| 2263 |
+
},
|
| 2264 |
+
{
|
| 2265 |
+
"epoch": 1.8098510882016037,
|
| 2266 |
+
"grad_norm": 11.285879135131836,
|
| 2267 |
+
"learning_rate": 6.641368552840088e-08,
|
| 2268 |
+
"loss": 0.7741,
|
| 2269 |
+
"step": 3160
|
| 2270 |
+
},
|
| 2271 |
+
{
|
| 2272 |
+
"epoch": 1.8155784650630011,
|
| 2273 |
+
"grad_norm": 8.553770065307617,
|
| 2274 |
+
"learning_rate": 6.250071798778117e-08,
|
| 2275 |
+
"loss": 0.7114,
|
| 2276 |
+
"step": 3170
|
| 2277 |
+
},
|
| 2278 |
+
{
|
| 2279 |
+
"epoch": 1.8213058419243986,
|
| 2280 |
+
"grad_norm": 8.297019004821777,
|
| 2281 |
+
"learning_rate": 5.870409765960966e-08,
|
| 2282 |
+
"loss": 0.761,
|
| 2283 |
+
"step": 3180
|
| 2284 |
+
},
|
| 2285 |
+
{
|
| 2286 |
+
"epoch": 1.827033218785796,
|
| 2287 |
+
"grad_norm": 9.877187728881836,
|
| 2288 |
+
"learning_rate": 5.502413183182592e-08,
|
| 2289 |
+
"loss": 0.7156,
|
| 2290 |
+
"step": 3190
|
| 2291 |
+
},
|
| 2292 |
+
{
|
| 2293 |
+
"epoch": 1.8327605956471937,
|
| 2294 |
+
"grad_norm": 11.764557838439941,
|
| 2295 |
+
"learning_rate": 5.146111835067646e-08,
|
| 2296 |
+
"loss": 0.6779,
|
| 2297 |
+
"step": 3200
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 1.8384879725085912,
|
| 2301 |
+
"grad_norm": 11.132621765136719,
|
| 2302 |
+
"learning_rate": 4.8015345596608805e-08,
|
| 2303 |
+
"loss": 0.7472,
|
| 2304 |
+
"step": 3210
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 1.8442153493699887,
|
| 2308 |
+
"grad_norm": 11.750364303588867,
|
| 2309 |
+
"learning_rate": 4.468709246093133e-08,
|
| 2310 |
+
"loss": 0.6553,
|
| 2311 |
+
"step": 3220
|
| 2312 |
+
},
|
| 2313 |
+
{
|
| 2314 |
+
"epoch": 1.8499427262313861,
|
| 2315 |
+
"grad_norm": 14.244782447814941,
|
| 2316 |
+
"learning_rate": 4.147662832323884e-08,
|
| 2317 |
+
"loss": 0.8058,
|
| 2318 |
+
"step": 3230
|
| 2319 |
+
},
|
| 2320 |
+
{
|
| 2321 |
+
"epoch": 1.8556701030927836,
|
| 2322 |
+
"grad_norm": 10.789249420166016,
|
| 2323 |
+
"learning_rate": 3.838421302961098e-08,
|
| 2324 |
+
"loss": 0.7955,
|
| 2325 |
+
"step": 3240
|
| 2326 |
+
},
|
| 2327 |
+
{
|
| 2328 |
+
"epoch": 1.861397479954181,
|
| 2329 |
+
"grad_norm": 9.885890007019043,
|
| 2330 |
+
"learning_rate": 3.541009687158109e-08,
|
| 2331 |
+
"loss": 0.6453,
|
| 2332 |
+
"step": 3250
|
| 2333 |
+
},
|
| 2334 |
+
{
|
| 2335 |
+
"epoch": 1.8671248568155785,
|
| 2336 |
+
"grad_norm": 9.926796913146973,
|
| 2337 |
+
"learning_rate": 3.255452056587754e-08,
|
| 2338 |
+
"loss": 0.7055,
|
| 2339 |
+
"step": 3260
|
| 2340 |
+
},
|
| 2341 |
+
{
|
| 2342 |
+
"epoch": 1.872852233676976,
|
| 2343 |
+
"grad_norm": 12.368999481201172,
|
| 2344 |
+
"learning_rate": 2.98177152349417e-08,
|
| 2345 |
+
"loss": 0.6359,
|
| 2346 |
+
"step": 3270
|
| 2347 |
+
},
|
| 2348 |
+
{
|
| 2349 |
+
"epoch": 1.8785796105383734,
|
| 2350 |
+
"grad_norm": 10.849148750305176,
|
| 2351 |
+
"learning_rate": 2.7199902388220343e-08,
|
| 2352 |
+
"loss": 0.6735,
|
| 2353 |
+
"step": 3280
|
| 2354 |
+
},
|
| 2355 |
+
{
|
| 2356 |
+
"epoch": 1.8843069873997709,
|
| 2357 |
+
"grad_norm": 12.963729858398438,
|
| 2358 |
+
"learning_rate": 2.47012939042387e-08,
|
| 2359 |
+
"loss": 0.6878,
|
| 2360 |
+
"step": 3290
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"epoch": 1.8900343642611683,
|
| 2364 |
+
"grad_norm": 10.135659217834473,
|
| 2365 |
+
"learning_rate": 2.2322092013450313e-08,
|
| 2366 |
+
"loss": 0.657,
|
| 2367 |
+
"step": 3300
|
| 2368 |
+
},
|
| 2369 |
+
{
|
| 2370 |
+
"epoch": 1.8957617411225658,
|
| 2371 |
+
"grad_norm": 16.199338912963867,
|
| 2372 |
+
"learning_rate": 2.0062489281870156e-08,
|
| 2373 |
+
"loss": 0.8076,
|
| 2374 |
+
"step": 3310
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"epoch": 1.9014891179839633,
|
| 2378 |
+
"grad_norm": 12.090662002563477,
|
| 2379 |
+
"learning_rate": 1.7922668595487744e-08,
|
| 2380 |
+
"loss": 0.7334,
|
| 2381 |
+
"step": 3320
|
| 2382 |
+
},
|
| 2383 |
+
{
|
| 2384 |
+
"epoch": 1.9072164948453607,
|
| 2385 |
+
"grad_norm": 10.486949920654297,
|
| 2386 |
+
"learning_rate": 1.5902803145465994e-08,
|
| 2387 |
+
"loss": 0.7883,
|
| 2388 |
+
"step": 3330
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"epoch": 1.9129438717067582,
|
| 2392 |
+
"grad_norm": 8.531800270080566,
|
| 2393 |
+
"learning_rate": 1.4003056414122583e-08,
|
| 2394 |
+
"loss": 0.6773,
|
| 2395 |
+
"step": 3340
|
| 2396 |
+
},
|
| 2397 |
+
{
|
| 2398 |
+
"epoch": 1.9186712485681556,
|
| 2399 |
+
"grad_norm": 10.624215126037598,
|
| 2400 |
+
"learning_rate": 1.222358216169922e-08,
|
| 2401 |
+
"loss": 0.6701,
|
| 2402 |
+
"step": 3350
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"epoch": 1.9243986254295533,
|
| 2406 |
+
"grad_norm": 12.858296394348145,
|
| 2407 |
+
"learning_rate": 1.0564524413915422e-08,
|
| 2408 |
+
"loss": 0.6576,
|
| 2409 |
+
"step": 3360
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 1.9301260022909508,
|
| 2413 |
+
"grad_norm": 10.357311248779297,
|
| 2414 |
+
"learning_rate": 9.026017450312673e-09,
|
| 2415 |
+
"loss": 0.5779,
|
| 2416 |
+
"step": 3370
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 1.9358533791523482,
|
| 2420 |
+
"grad_norm": 8.941463470458984,
|
| 2421 |
+
"learning_rate": 7.608185793385458e-09,
|
| 2422 |
+
"loss": 0.7369,
|
| 2423 |
+
"step": 3380
|
| 2424 |
+
},
|
| 2425 |
+
{
|
| 2426 |
+
"epoch": 1.9415807560137457,
|
| 2427 |
+
"grad_norm": 8.527290344238281,
|
| 2428 |
+
"learning_rate": 6.311144198503316e-09,
|
| 2429 |
+
"loss": 0.6539,
|
| 2430 |
+
"step": 3390
|
| 2431 |
+
},
|
| 2432 |
+
{
|
| 2433 |
+
"epoch": 1.9473081328751431,
|
| 2434 |
+
"grad_norm": 9.527642250061035,
|
| 2435 |
+
"learning_rate": 5.134997644622108e-09,
|
| 2436 |
+
"loss": 0.6612,
|
| 2437 |
+
"step": 3400
|
| 2438 |
+
},
|
| 2439 |
+
{
|
| 2440 |
+
"epoch": 1.9530355097365406,
|
| 2441 |
+
"grad_norm": 9.823323249816895,
|
| 2442 |
+
"learning_rate": 4.079841325787981e-09,
|
| 2443 |
+
"loss": 0.7576,
|
| 2444 |
+
"step": 3410
|
| 2445 |
+
},
|
| 2446 |
+
{
|
| 2447 |
+
"epoch": 1.9587628865979383,
|
| 2448 |
+
"grad_norm": 12.560805320739746,
|
| 2449 |
+
"learning_rate": 3.145760643432527e-09,
|
| 2450 |
+
"loss": 0.6648,
|
| 2451 |
+
"step": 3420
|
| 2452 |
+
},
|
| 2453 |
+
{
|
| 2454 |
+
"epoch": 1.9644902634593358,
|
| 2455 |
+
"grad_norm": 11.428935050964355,
|
| 2456 |
+
"learning_rate": 2.33283119946015e-09,
|
| 2457 |
+
"loss": 0.7002,
|
| 2458 |
+
"step": 3430
|
| 2459 |
+
},
|
| 2460 |
+
{
|
| 2461 |
+
"epoch": 1.9702176403207332,
|
| 2462 |
+
"grad_norm": 9.424127578735352,
|
| 2463 |
+
"learning_rate": 1.6411187901297919e-09,
|
| 2464 |
+
"loss": 0.6452,
|
| 2465 |
+
"step": 3440
|
| 2466 |
+
},
|
| 2467 |
+
{
|
| 2468 |
+
"epoch": 1.9759450171821307,
|
| 2469 |
+
"grad_norm": 10.48491096496582,
|
| 2470 |
+
"learning_rate": 1.0706794007290243e-09,
|
| 2471 |
+
"loss": 0.5779,
|
| 2472 |
+
"step": 3450
|
| 2473 |
+
},
|
| 2474 |
+
{
|
| 2475 |
+
"epoch": 1.9816723940435281,
|
| 2476 |
+
"grad_norm": 9.892742156982422,
|
| 2477 |
+
"learning_rate": 6.215592010426763e-10,
|
| 2478 |
+
"loss": 0.7057,
|
| 2479 |
+
"step": 3460
|
| 2480 |
+
},
|
| 2481 |
+
{
|
| 2482 |
+
"epoch": 1.9873997709049256,
|
| 2483 |
+
"grad_norm": 10.53469181060791,
|
| 2484 |
+
"learning_rate": 2.937945416166543e-10,
|
| 2485 |
+
"loss": 0.684,
|
| 2486 |
+
"step": 3470
|
| 2487 |
+
},
|
| 2488 |
+
{
|
| 2489 |
+
"epoch": 1.993127147766323,
|
| 2490 |
+
"grad_norm": 10.819581031799316,
|
| 2491 |
+
"learning_rate": 8.741195081479747e-11,
|
| 2492 |
+
"loss": 0.6306,
|
| 2493 |
+
"step": 3480
|
| 2494 |
+
},
|
| 2495 |
+
{
|
| 2496 |
+
"epoch": 1.9988545246277205,
|
| 2497 |
+
"grad_norm": 13.76382064819336,
|
| 2498 |
+
"learning_rate": 2.428132672926964e-12,
|
| 2499 |
+
"loss": 0.6657,
|
| 2500 |
+
"step": 3490
|
| 2501 |
+
},
|
| 2502 |
+
{
|
| 2503 |
+
"epoch": 2.0,
|
| 2504 |
+
"step": 3492,
|
| 2505 |
+
"total_flos": 8.351047625159475e+16,
|
| 2506 |
+
"train_loss": 0.9289274918937465,
|
| 2507 |
+
"train_runtime": 3314.9598,
|
| 2508 |
+
"train_samples_per_second": 4.212,
|
| 2509 |
+
"train_steps_per_second": 1.053
|
| 2510 |
+
}
|
| 2511 |
+
],
|
| 2512 |
+
"logging_steps": 10,
|
| 2513 |
+
"max_steps": 3492,
|
| 2514 |
+
"num_input_tokens_seen": 0,
|
| 2515 |
+
"num_train_epochs": 2,
|
| 2516 |
+
"save_steps": 500,
|
| 2517 |
+
"stateful_callbacks": {
|
| 2518 |
+
"TrainerControl": {
|
| 2519 |
+
"args": {
|
| 2520 |
+
"should_epoch_stop": false,
|
| 2521 |
+
"should_evaluate": false,
|
| 2522 |
+
"should_log": false,
|
| 2523 |
+
"should_save": true,
|
| 2524 |
+
"should_training_stop": true
|
| 2525 |
+
},
|
| 2526 |
+
"attributes": {}
|
| 2527 |
+
}
|
| 2528 |
+
},
|
| 2529 |
+
"total_flos": 8.351047625159475e+16,
|
| 2530 |
+
"train_batch_size": 1,
|
| 2531 |
+
"trial_name": null,
|
| 2532 |
+
"trial_params": null
|
| 2533 |
+
}
|